Defending the Integrity of Machine Learning Models in the Face of Adversarial Threats

Defending the Integrity of Machine Learning Models in the Face of Adversarial Threats | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Defending the Integrity of Machine Learning Models in the Face of Adversarial Threats Hrittick Roy, Sourabh Anand This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3871878/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract The primary objective of the research is to examine the resilience of models based on machine learning when confronted with adversarial threats. The focus will be on developing effective strategies to safeguard the integrity of these models. This study employs a carefully designed framework to undertake the task of simulating adversarial attacks on machine learning models. The central objective of the research centres on the intentional manipulation of numerical variables in order to accurately simulate real-world threats. The investigation of statistical analyses, such as t-tests and comprehensive data visualisations, enables the evaluation of the impact exerted by adversarial attacks on both data distributions and model outputs. The primary objective of the current investigation is to assess the effectiveness of a mimicked defence mechanism in the domain of machine learning model restoration, with the ultimate goal of preserving model integrity. The results gathered by our study emphasise the significant significance of adversarial threats, thereby underscoring the need for a comprehensive analysis of robust defence strategies. The analysis of correlation matrices, both before and after adversarial attacks, provides valuable insights into the changes in the fundamental structure induced by these interventions. The current investigation offers valuable insights into the preservation of the integrity of machine learning models and suggests possible ways to enhance their resilience against ever-changing adversarial tactics. Adversarial Attacks Artificial Intelligence Security Cryptography Secure Federated Learning Model Forensics Introduction Machine learning models plays a very vital and really super important part in various and many domains, including decision-making processes and predictive analytics. As the prevalence and popularity of these models experiences an ongoing increase and rise, there is a concomitant and really connected amplification and increase in the susceptibility and vulnerabilities to adversarial threats that may undermine and really destroy their integrity and trust that they have gained. The exploration and studying of strategies and methods aimed at bolstering and making the machine learning models more robust and super tough and strong in the face of adversarial attacks has come to be very crucial and urgent and like a super important area and field for inquiry and examination. The primary and first major objective and aim of the present investigation is to examine and look into the critical and super very important issue and problem of preserving the reliability of machine learning models when confronted with adversarial threats and dangers and hazards. The central and main focus and point of this investigation pertains to the augmentation and improving and maximizing of predictive machine learning models in order to maintain and hold and keep their high level of accuracy and reliability when confronted and faced with deliberate adversarial manipulations and changes. The comprehension and understanding and mitigation and dealing with these potential risks and dangers and threats and hazards are of like super very significant and important and play a big role in upholding and maintaining the integrity and dependability and trustworthiness of machine learning applications across various and many different domains. The current and latest investigation utilised and made use of a really diverse range and variety of methodologies and methods to explore and study the research inquiry and question under and in consideration and thought. The methods used and employed in the present investigation were conducted and carried out with great care and attention and in order to thoroughly and very completely examine and study the research question at hand. A meticulously and very carefully and painstakingly designed framework and system was utilised and used to replicate and make copies of and also analyse and study adversarial attacks on machine learning models. The current and latest framework prioritises and gives priority to the manipulation and changing of numerical variables to effectively and very effectively replicate and copy real-world threats and dangers. The current and ongoing investigation utilises and makes use of a range and variety of statistical analyses and computations, including t-tests, in conjunction and along with thorough data visualisations, in order to shed light and illuminate and bring into focus the impacts and effects of adversarial attacks on both data distributions and model outputs. Our research analysis has and have yielded and given rise to the following and the really next key findings and discoveries and revelations and findings: Our analysis reveals and shows and proves a noteworthy and very important correlation and association and connection between the X and Y variables. The findings of our study have and has unveiled and exposed a strong and very thick positive correlation and connection and association, highlighted and emphasized the significance and importance and value of adversarial threats and dangers and provided insights and knowledge into the basic weaknesses and shortcomings that exist and are present and can be found within machine learning models. The primary and first major and very chief and principal objective and aim of this study is to introduce and bring forward a newly developed and created simulated defence mechanism and evaluate and judge and assess and analyse and think about its efficacy and effectiveness and power and strength in the specific domain and field of model restoration and bringing back and restoring to the initial condition subsequent and after an attack and assault. Upon conducting and doing an initial and first analysis and examination and study of the correlation matrices, it is apparent and very clear and obvious and evident and easy to understand and see that there are discernible and noticeable structural modifications and changes arising from the adversarial interventions and invasions and intrusions. The previously and earlier mentioned and talked about observation and note remains and stays consistent and the same for both the correlation matrices observed and examined and looked at prior to the attacks and those observed and studied and looked at subsequent and following to the attacks. The aforementioned and before mentioned findings and discoveries offer and give significant and valuable insights and knowledge and understanding into the effects and impacts and consequences of adversarial interventions and intrusions on the fundamental and basic structure and foundation of the data. The introductory and beginning section adeptly and very perfectly establishes and sets up the necessary and essential framework and structure and foundation for conducting and doing a comprehensive and full exploration and investigation into the preservation and looking after and taking care of the validity and truthfulness and genuineness and authenticity of machine learning models. The provided statement highlights and underlines and stresses and emphasizes the importance and significance and value of the research investigation and also provides a short and brief and concise and quick overview and summary of the methodology and notable and distinguished findings and discoveries that will be subsequently and afterwards presented and shown and displayed and offered.. 1.1. Related work Ishai Rosenberg et.al.[1] The article investigates machine learning vulnerability in cybersecurity due to adversarial attacks, categorizing methods and applications. Pioneering a unified taxonomy, it addresses challenges in implementing end-to-end adversarial attacks and guides future cybersecurity research. Mingfu Xue et.al.[2] the vulnerability of machine learning models to diverse attacks, covering training set poisoning, adversarial examples, and more. It systematically analyses threat models, attack methods, and defence strategies, offering real-world attack instances and recommendations for security evaluations. The paper concludes by outlining future directions to enhance machine learning security. Qiang Liu et.al.[3] This paper emphasizes machine learning's broad application while addressing the vulnerability of algorithms and training data to security threats. It surveys threats against learning algorithms, categorizing them into training and testing phases, and systematically classifies defensive techniques, paving the way for future studies on emerging trends in machine learning security. Anirban Chakraborty et.al.[4] Deep learning, a potent framework for complex tasks, faces security vulnerabilities from imperceptible adversarial examples, risking misclassification. Adversaries exploit these weaknesses with diverse threat models, underscoring the need for robust countermeasures. This paper succinctly explores adversarial attacks, their threat models, and evaluates recent countermeasures for enhancing the security of deep learning systems. Xiaoyong Yuan et.al.[5] Deep learning, despite its success, faces a critical challenge with imperceptible adversarial examples that jeopardize safety-critical applications. This paper reviews adversarial examples, categorizes generation methods, and discusses countermeasures, emphasizing the imperative need for robust defences in safeguarding deep learning applications. Wieland Brendel et.al.[6] This study addresses the vulnerability of machine learning algorithms to imperceptible perturbations, emphasizing the risk in real-world applications. It introduces the Boundary Attack, a decision-based method applicable to black-box models, requiring minimal hyperparameter tuning and outperforming gradient-based attacks in tasks like ImageNet. The findings highlight the importance of studying model robustness and raise safety concerns for deployed machine learning systems. Guofu Li et.al.[7] Adversarial machine learning investigates intentional data synthesis to mislead well-trained models, with a recent focus on imperceptible perturbations impacting deep learning accuracy. This paper offers a comprehensive overview of the field, covering foundations, attacking and defending strategies, and recent developments in deep learning's susceptibility to adversarial attacks. Dou An et.al.[8] This paper tackles data integrity threats in smart grids through a Deep-Q-Network Detection (DQND) scheme, utilizing deep reinforcement learning for enhanced defense. DQND, overcoming dimensionality challenges, employs main and target networks for optimal defense strategy learning, displaying improved accuracy and speed in experimental evaluations on 9, 14, and 30 bus systems. Lingjuan Lyu et.al.[9] This article explores federated learning (FL) as a privacy-preserving alternative to centralized AI model training, highlighting the vulnerabilities of existing FL protocols to adversaries. The survey covers threat models, privacy attacks, defences, and poisoning attacks, providing insights into the evolving landscape of FL security over the past five years. It emphasizes the need for robust and privacy-ensuring FL systems in the face of diverse challenges. Akhil Goel et.al.[10] This paper introduces a novel defence mechanism, a "defence layer," to safeguard deep learning models from adversarial attacks, particularly in black-box and Gray-box settings. The parameter-free defence layer, applicable to various convolutional networks (CNNs), proves effective against attacks like FGSM, L_2, Elastic-Net, and Deep Fool. Experimental results on databases (MNIST, CIFAR-10, PaSC) and CNN architectures (VGG, ResNet, Dense Net) demonstrate the efficacy of the proposed defence layer without introducing computational overhead. For instance, on CIFAR-10, while an attack reduces ResNet-50 accuracy to 6.3%, the "defines layer" maintains an accuracy of 81.32%. Methodology The data subsequent generations process encompasses the systematic acquisition and generation of data with the specific objective of facilitating research endeavours. The implementation of this task can be achieved through a diverse range of methodologies, including but not limited to the utilisation of surveys and experimentation techniques. The synthetic dataset employed in this research endeavour has been carefully crafted to closely emulate and reproduce the complex dynamics and characteristics observed in real-world scenarios. The numerical variables, denoted as to are obtained through the utilisation of diverse probability distributions to encompass an extensive range of data characteristics. The variables being examined in this study are referred to as variable1 and variable2. The data points were obtained through a sampling process from a standard normal distribution , which is a widely employed statistical model for representing continuous features that demonstrate characteristic behaviour. A categorical variable consisting of two classes, denoted as 'A' and 'B', was randomly generated for the reason of introducing a categorical dimension into the research analysis. Variable3 is a crucial component of the research study being conducted. The identified variable demonstrates potential significance and relevance to the values that were sampled from a distribution that is uniform, specifically the distribution. This sampling technique was employed to introduce variability within a predetermined range. The term "Variable4" is commonly employed in the realm of research to refer to a particular variable that is under investigation within a specific study or experiment. It is of utmost significance to acknowledge that the simulation of a continuous variable was carried out through the utilisation of a normal distribution characterised by a mean value of zero and a standard deviation of one, conventionally represented as.The primary objective of the current investigation is to examine the phenomenon of simulating adversarial attacks. Adversarial attacks encompass intentional endeavours aimed at manipulating machine learning models through the introduction of meticulously crafted input data. In order to simulate adversarial threats, the researchers implemented deliberate attacks by introducing perturbations to a specific subset of the dataset. In the present inquiries, a comprehensive set of 20 samples was meticulously chosen through the implementation of a random sampling technique. The selection of these data points was conducted with the explicit objective of manipulating a range of numerical variables. The attack is effectively depicted and measured by employing the following mathematical expression: The attacked value can be computed by adding the perturbation to the original value. The calculation of the attacked value involves the summation of the original value and the perturbation. The deliberate perturbation utilised within the scope of this research endeavour aims to replicate scenarios wherein an individual with malicious intent manipulates input data in order to deceive machine learning models. The statistical analyses performed in the present study were undertaken with the objective of investigating the associations and trends present within the gathered data. The current investigation utilised a range of statistical techniques to examine the data. Specifically, the emphasis of this evaluation is on the implementation of descriptive statistics. Descriptive statistics, as a fundamental branch of statistical analysis, endeavours to succinctly summarise and elucidate the primary attributes and features inherent within a given dataset. The calculation of essential statistical metrics, such as the mean standard deviation and quartiles, is conducted for each numerical variable both before and after the adversarial attack. The t-test is a widely employed statistical test that is utilised to assess whether there exists a statistically significant distinction between the means of two groups or samples. A comparative analysis using a two-sample t-test was performed to assess the level of statistical significance in the differences observed in Variable1 between the 'A' and 'B' categories, both before and after the occurrence of the attack. The null hypothesis, commonly represented as , is formulated based on the underlying premise that there exists no statistically significant distinction between the variables under investigation. In contrast, the alternative hypothesis, commonly represented as , posits the presence of a substantial disparity between the variables under investigation. The employment of data visualisations is a widely adopted methodology across diverse domains, encompassing but not restricted to, data analysis, business intelligence, and scientific research. Boxplots are a commonly employed graphical tool for the purpose of visually examining and contrasting the distribution of numerical variables across different categories, both before and after the onset of an adversarial attack. Histograms provide researchers with valuable insights into the frequency distribution of variables prior to the start of an attack. The primary objective of these statistical examinations is to elucidate the influence of adversarial attacks on the basic characteristics of data and the outcomes produced by models. The aforementioned statement provides a fundamental basis for future discussions regarding the development of strategies aimed at preventing similar attacks and improving the robustness of models. Result The present study's findings provide valuable insights into the impact of adversarial attacks on machine learning models and the success rate of various defensive strategies. The study's results can be categorised into two main components: descriptive stats and T-test Analysis. Descriptive statistics, as a fundamental branch of statistical analysis, encompasses the methodical gathering, arrangement, examination, and elucidation of data. The objective of this study is to provide a summary and description of the descriptive statistical data that were computed for each numerical variable, prior to as well as after the adversarial attack took place. Table 1 provides a succinct overview of the key statistical measures, including the mean deviation from the mean and quartiles. The tabular representation presented herein provides an extensive and informative representation of the fundamental characteristics inherent within the dataset under consideration. The correlation matrices that were observed offer a visual depiction of the alterations in the associations among parameters subsequent to the execution of the adversarial attack. Further investigation is imperative in order to elucidate the key variables that contribute to the previously mentioned changes and to comprehend their potential implications on the effectiveness of the model. The presented study's findings provide significant benefits to the comprehension of the impacts of adversarial attacks on data distribution and the intricate relationship among variables. The aforementioned insights provide a solid basis for subsequent deliberations and inquiries pertaining to the robustness of models and the formulation of efficacious defence tactics . Discussion The focal point of the ongoing discourse is all about the subject matter of information for subsequent generations and attack simulation within the framework of code development. The specific aspect under consideration holds considerable significance within the realm of cybersecurity research and development. The objective of this discourse is to thoroughly examine the complexities associated with data generation as well as attack computer simulation methodologies, with a focus on their significance in evaluating the resilience and efficacy of safety precautions integrated within software code. The term "data generation" refers to the systematic procedure of generating synthetic data sets. In the current code implementation, a synthetic dataset is being created, consisting of five numerical variables that are labelled as Variable1 to Variable4, in addition to a categorical variable known as Category. Numerical variables are commonly derived from a diverse set of distributions order to accurately simulate and mimic real-world scenarios. The variable 'Category' consists of two discrete classes, denoted as 'A' and 'B', which have been randomly assigned. The process of simulating an adversarial attack involves intentionally introducing perturbations to a targeted subset within the dataset. For the purpose of this study, a sample size of twenty random samples was selected. The numerical variables present in each sample underwent a uniform increment by a constant value, specifically 5 units. The present scenario involves the execution of a simulated adversarial manipulation of input data, with the explicit objective of deceiving machine learning models. The code provided implements a t-test to assess the statistical significance of the differences in Variable1 values between categories A and B. This analysis conducted separately for the pre-attack and post-attack periods. The generation of boxplots and histograms serves the purpose of visually illustrating the distribution of each numerical variable across different categories prior to the onset of the attack. The preceding assertion provides significant observations regarding the attributes of the data being examined. The implementation of the defence mechanism involves the removal of samples that have been affected. Following the attack and subsequent defence, a series of new boxplots were generated to analyse the resulting changes in the distribution of the data. The presentation of summary statistics serves the purpose of providing a comprehensive overview of the dataset, encompassing key measures that encapsulate its central tendencies and variability. The current investigation involves the calculation and subsequent presentation of the correlation matrix before and after the attack event. Heatmaps have emerged as a highly valuable and indispensable tool within the domain of data visualization. They provide researchers with a powerful means to accurately represent and analyse the complex interdependencies that exist among different variables. By employing colour gradients to represent the intensity or magnitude of a particular variable, heatmaps offer a visually compelling way to uncover patterns, trends, and relationships that may otherwise remain hidden in raw data. This enables researchers to gain deeper insights and make informed decisions based on a comprehensive understanding of the data at hand. The examination of the effects of adversarial attacks on the interconnectedness of numerical variables holds significant significance in the pursuit of attaining a holistic comprehension of this phenomenon. The analysis of the findings involves leveraging the t-test results and visual depictions to acquire a deeper understanding of the impact of the adversarial attack on the data's attributes. The utilization of visual representations, such as boxplots and histograms, offers a comprehensive portrayal of the influence exerted by the attack on the distribution of numerical variables. The graphical displays employed in this study effectively communicate any alterations or deviations in the original distribution resulting from the attack. Moreover, the correlation matrices provide valuable insights into the changes observed in the associations among variables. Through a meticulous analysis of the correlation coefficients, it is possible to discern any discernible shifts or delays in the associations between various variables that have arisen as a consequence of the attack. The investigation delves into the ramifications of climate change on ecosystems at a global scale. The present inquiry delves into the pivotal role of genetic modification in the realm of agriculture. The examination of ethical considerations surrounding the utilization of animal testing in scientific research is a topic of great significance. The ethical implications of subjecting animals to experimentation for the advancement of scientific knowledge have been a subject of intense debate and scrutiny. This discourse revolves around the moral obligations and responsibilities that researchers and society at large bear towards animals, as well as the potential benefits and drawbacks associated with such practices. By delving into the ethical dimensions of animal testing, researchers aim to the investigation and exploration of the efficacy of adversarial attacks have garnered significant attention within the domains of machine learning and computer vision. The investigation into the exploitative capabilities of these attacks on deep learning models, with the intention of deceiving them and exerting influence, has been a subject of interest among researchers. Specifically, the impact of the adversarial attack on the statistically significant nature of Variable1 between categories A and B is deemed worthy of further examination and discussion. The investigation necessitates a comprehensive evaluation to determine the extent to which the attack successfully manipulated the data. Evaluation of defence Mechanisms: In this analysis, we will be examining the various defence mechanisms employed by individuals in order to cope with psychological stressors. Defence mechanisms are unconscious psychological strategies The evaluation of the simulated defence mechanism involves a thorough examination of the changes in data distribution that occur after the removal of the affected samples. Within the scope of this discussion, we shall undertake a comprehensive examination of the numerous constraints that have been identified within the given context, while also endeavouring to explore potential pathways for enhancement. In order to achieve a thorough comprehension of the potential obstacles that may impede the effectiveness or relevance of the impact analysis on data characteristics, it is essential to undertake a meticulous evaluation of these limitations. This examination can be facilitated by employing diverse statistical measures, as well as graphical representations like boxplots and histograms, which allow for the observation and interpretation of the attack's influence. The utilization of analytical tools facilitates the acquisition of valuable insights pertaining to alterations in the distribution and general statistics of the data. Summary statistics provide a succinct representation of key characteristics of a dataset, such as its central tendency, dispersion, and skewness. These measures, including the mean, median, standard deviation, and quartiles, offer valuable insights into the distribution and variability of the data. By examining these summary statistics, researchers can gain a comprehensive understanding of the dataset's overall pattern and characteristics. Through a comparative analysis of the summary statistics prior to and subsequent to the occurrence of the attack, it is possible to discern any noteworthy modifications in these quantitative measures. In the context of data analysis, it is worth noting that a shift in the mean or an increase in the standard deviation can potentially signify a disturbance in the underlying pattern of the data. Boxplots have been widely recognized as a valuable tool in data visualization, offering researchers a comprehensive understanding of data distribution and facilitating the identification of any potential outliers. Through a careful analysis of the boxplots both prior to and subsequent to the occurrence of the attack, we are able to evaluate any potential alterations in the dispersion, asymmetry, or existence of outliers. The current analysis focuses on investigating any noteworthy deviations within the dataset. Specifically, our objective is to explore the alterations in the relationship between variables by examining the changes that have been noticed in the correlation matrices. The careful consideration of the implications pertaining to machine learning algorithms that rely on these relationships is of utmost importance. Real-world applications refer to the practical utilization of a concept, theory, or technology in various domains outside of academic or theoretical settings. These applications the current investigation establishes a fundamental correlation between the acquired results and their pragmatic ramifications. The understanding and mitigation of adversarial attacks in the field of machine learning are of utmost importance. Conclusion In this study has explored the complex dynamics associated with adversarial attacks targeting machine learning models. Specifically, it has examined the effects of these attacks on the characteristics of the data and the subsequent countermeasures employed for defence. The present study provides a comprehensive analysis of the key findings and their corresponding implications. The following section presents a concise summary of these findings and their potential significance in the broader research context. The effects of adversarial attacks were examined in this study. The key findings indicate that these attacks can have significant impacts on the performance and robustness of machine learning models. Specifically, it was observed that adversarial attacks can lead to a decrease in accuracy and the findings of the conducted simulated adversarial attack, as indicated by the statistical analysis using t-test, revealed a noticeable and statistically significant effect on Variable1 when comparing categories, A and B. The aforementioned observation highlights the susceptibility of machine learning models to deliberate manipulations of data. The phenomenon under investigation pertains to shifts in data distribution. The examination of visual representations, specifically boxplots and histograms, provided valuable insights into the notable alterations in the distribution patterns of numerical variables both prior to and subsequent to the occurrence of the adversarial attack. The aforementioned observation underscores the possibility of feature representations being distorted, which in turn can significantly impact the decision-making process of a model. The present study aims to conduct a comprehensive assessment of defence mechanisms. Defense mechanisms are psychological strategies employed by individuals to cope with various internal and external stressors. Understanding these mechanisms can provide the observed defence mechanism, which involved the removal of affected samples, exhibited a certain level of efficacy in the restoration of data integrity. Further investigation and refinement of defence strategies are necessary in order to improve their resilience and effectiveness. The present study aims to investigate potential alterations in the correlation matrix. The examination of correlation matrices has unveiled notable alterations in the associations among numerical variables subsequent to the implementation of attack and defence strategies. Comprehending these modifications is of utmost importance in evaluating the dependability of model forecasts when confronted with adversarial manipulations. The implications of the aforementioned findings are of significant importance and warrant further investigation. The results suggest potential ramifications The investigation of real-world model resilience is a topic of great significance in the field of research. The examination of how models perform and adapt in real-world scenarios is crucial for understanding the present study highlights the practical implications of adversarial attacks, thereby emphasizing the crucial need for robust defence mechanisms in order to protect machine learning models in real-world scenarios. Data quality assurance is a crucial aspect of any research study or data analysis process. It involves ensuring the accuracy, reliability, and consistency of the data being used. By implementing various quality, the examination of adversarial attacks and their effects on data distribution sheds light on the necessity of continuous endeavours in guaranteeing the quality and integrity of data, especially in situations where machine learning models hold significant importance. The user's text is concise and lacks specific information regarding the contributions made to a particular field. Further elaboration is required to provide a comprehensive understanding of the contributions made by individuals or groups. The present study makes a valuable contribution to the ever-changing field of machine learning security by conducting a comprehensive investigation into the ramifications of adversarial attacks on both data and model outputs. The findings presented in this study provide valuable insights for researchers, practitioners, and policymakers who are interested in enhancing the resilience of machine learning systems against deliberate manipulations. In summary, this research study provides a foundational basis for future investigations into the comprehension of adversarial threats and the formulation of advanced defence tactics to enhance the robustness of machine learning models in dynamic and arduous settings. Declarations Author Contribution Implemented the adversarial attack simulation, generated example data, and conducted statistical tests. References Rosenberg, I., Shabtai, A., Elovici, Y., & Rokach, L. (2021). Adversarial machine learning attacks and defense methods in the cyber security domain. ACM Computing Surveys (CSUR), 54 (5), 1–36. Xue, M., Yuan, C., Wu, H., Zhang, Y., & Liu, W. (2020). Machine learning security: Threats, countermeasures, and evaluations. IEEE Access, 8, 74720–74742. Liu, Q., Li, P., Zhao, W., Cai, W., Yu, S., & Leung, V. C. (2018). A survey on security threats and defensive techniques of machine learning: A data driven view. IEEE access, 6, 12103–12117. Chakraborty, A., Alam, M., Dey, V., Chattopadhyay, A., & Mukhopadhyay, D. (2018). Adversarial attacks and defences: A survey. arXiv preprint arXiv:1810.00069. Yuan, X., He, P., Zhu, Q., & Li, X. (2019). Adversarial examples: Attacks and defenses for deep learning. IEEE transactions on neural networks and learning systems, 30(9), 2805–2824. Brendel, W., Rauber, J., & Bethge, M. (2017). Decision-based adversarial attacks: Reliable attacks against black-box machine learning models. arXiv preprint arXiv:1712.04248. Li, G., Zhu, P., Li, J., Yang, Z., Cao, N., & Chen, Z. (2018). Security matters: A survey on adversarial machine learning. arXiv preprint arXiv:1810.07339. An, D., Yang, Q., Liu, W., & Zhang, Y. (2019). Defending against data integrity attacks in smart grid: A deep reinforcement learning-based approach. IEEE Access, 7, 110835–110845. Lyu, L., Yu, H., Ma, X., Chen, C., Sun, L., Zhao, J., ... & Philip, S. Y. (2022). Privacy and robustness in federated learning: Attacks and defenses. IEEE transactions on neural networks and learning systems. Goel, A., Agarwal, A., Vatsa, M., Singh, R., & Ratha, N. K. (2020). DNDNet: Reconfiguring CNN for adversarial robustness. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 22–23). Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3871878","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":267647011,"identity":"c0c66b95-f0b4-4917-beeb-5d066e27eb03","order_by":0,"name":"Hrittick Roy","email":"data:image/png;base64,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","orcid":"","institution":"University School of Information, Communication and Technology Guru Gobind Singh Indraprastha University","correspondingAuthor":true,"prefix":"","firstName":"Hrittick","middleName":"","lastName":"Roy","suffix":""},{"id":267647012,"identity":"97b1ebe8-b0b6-43f2-8d6c-7d94fd9d3820","order_by":1,"name":"Sourabh Anand","email":"","orcid":"","institution":"University School of Information, Communication and Technology Guru Gobind Singh Indraprastha University","correspondingAuthor":false,"prefix":"","firstName":"Sourabh","middleName":"","lastName":"Anand","suffix":""}],"badges":[],"createdAt":"2024-01-17 04:59:14","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3871878/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3871878/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":50985870,"identity":"9fce9e89-0f0f-4a18-a544-ba4b6647dbe2","added_by":"auto","created_at":"2024-02-12 08:00:04","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":376608,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3871878/v1/6d006d0b-e50f-4d53-967f-f52c0a59f84e.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Defending the Integrity of Machine Learning Models in the Face of Adversarial Threats","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMachine learning models plays a very vital and really super important part in various and many domains, including decision-making processes and predictive analytics. As the prevalence and popularity of these models experiences an ongoing increase and rise, there is a concomitant and really connected amplification and increase in the susceptibility and vulnerabilities to adversarial threats that may undermine and really destroy their integrity and trust that they have gained. The exploration and studying of strategies and methods aimed at bolstering and making the machine learning models more robust and super tough and strong in the face of adversarial attacks has come to be very crucial and urgent and like a super important area and field for inquiry and examination. The primary and first major objective and aim of the present investigation is to examine and look into the critical and super very important issue and problem of preserving the reliability of machine learning models when confronted with adversarial threats and dangers and hazards. The central and main focus and point of this investigation pertains to the augmentation and improving and maximizing of predictive machine learning models in order to maintain and hold and keep their high level of accuracy and reliability when confronted and faced with deliberate adversarial manipulations and changes. The comprehension and understanding and mitigation and dealing with these potential risks and dangers and threats and hazards are of like super very significant and important and play a big role in upholding and maintaining the integrity and dependability and trustworthiness of machine learning applications across various and many different domains. The current and latest investigation utilised and made use of a really diverse range and variety of methodologies and methods to explore and study the research inquiry and question under and in consideration and thought. The methods used and employed in the present investigation were conducted and carried out with great care and attention and in order to thoroughly and very completely examine and study the research question at hand. A meticulously and very carefully and painstakingly designed framework and system was utilised and used to replicate and make copies of and also analyse and study adversarial attacks on machine learning models. The current and latest framework prioritises and gives priority to the manipulation and changing of numerical variables to effectively and very effectively replicate and copy real-world threats and dangers. The current and ongoing investigation utilises and makes use of a range and variety of statistical analyses and computations, including t-tests, in conjunction and along with thorough data visualisations, in order to shed light and illuminate and bring into focus the impacts and effects of adversarial attacks on both data distributions and model outputs. Our research analysis has and have yielded and given rise to the following and the really next key findings and discoveries and revelations and findings: Our analysis reveals and shows and proves a noteworthy and very important correlation and association and connection between the X and Y variables. The findings of our study have and has unveiled and exposed a strong and very thick positive correlation and connection and association, highlighted and emphasized the significance and importance and value of adversarial threats and dangers and provided insights and knowledge into the basic weaknesses and shortcomings that exist and are present and can be found within machine learning models. The primary and first major and very chief and principal objective and aim of this study is to introduce and bring forward a newly developed and created simulated defence mechanism and evaluate and judge and assess and analyse and think about its efficacy and effectiveness and power and strength in the specific domain and field of model restoration and bringing back and restoring to the initial condition subsequent and after an attack and assault. Upon conducting and doing an initial and first analysis and examination and study of the correlation matrices, it is apparent and very clear and obvious and evident and easy to understand and see that there are discernible and noticeable structural modifications and changes arising from the adversarial interventions and invasions and intrusions. The previously and earlier mentioned and talked about observation and note remains and stays consistent and the same for both the correlation matrices observed and examined and looked at prior to the attacks and those observed and studied and looked at subsequent and following to the attacks. The aforementioned and before mentioned findings and discoveries offer and give significant and valuable insights and knowledge and understanding into the effects and impacts and consequences of adversarial interventions and intrusions on the fundamental and basic structure and foundation of the data. The introductory and beginning section adeptly and very perfectly establishes and sets up the necessary and essential framework and structure and foundation for conducting and doing a comprehensive and full exploration and investigation into the preservation and looking after and taking care of the validity and truthfulness and genuineness and authenticity of machine learning models. The provided statement highlights and underlines and stresses and emphasizes the importance and significance and value of the research investigation and also provides a short and brief and concise and quick overview and summary of the methodology and notable and distinguished findings and discoveries that will be subsequently and afterwards presented and shown and displayed and offered..\u003c/p\u003e\n\u003cp\u003e1.1. Related work\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIshai Rosenberg et.al.[1]\u0026nbsp;The article investigates machine learning vulnerability in cybersecurity due to adversarial attacks, categorizing methods and applications. Pioneering a unified taxonomy, it addresses challenges in implementing end-to-end adversarial attacks and guides future cybersecurity research.\u003c/p\u003e\n\u003cp\u003eMingfu Xue et.al.[2] the vulnerability of machine learning models to diverse attacks, covering training set poisoning, adversarial examples, and more. It systematically analyses threat models, attack methods, and defence strategies, offering real-world attack instances and recommendations for security evaluations. The paper concludes by outlining future directions to enhance machine learning security.\u003c/p\u003e\n\u003cp\u003eQiang Liu et.al.[3] This paper emphasizes machine learning\u0026apos;s broad application while addressing the vulnerability of algorithms and training data to security threats. It surveys threats against learning algorithms, categorizing them into training and testing phases, and systematically classifies defensive techniques, paving the way for future studies on emerging trends in machine learning security.\u003c/p\u003e\n\u003cp\u003eAnirban Chakraborty et.al.[4] Deep learning, a potent framework for complex tasks, faces security vulnerabilities from imperceptible adversarial examples, risking misclassification. Adversaries exploit these weaknesses with diverse threat models, underscoring the need for robust countermeasures. This paper succinctly explores adversarial attacks, their threat models, and evaluates recent countermeasures for enhancing the security of deep learning systems.\u003c/p\u003e\n\u003cp\u003eXiaoyong Yuan et.al.[5] Deep learning, despite its success, faces a critical challenge with imperceptible adversarial examples that jeopardize safety-critical applications. This paper reviews adversarial examples, categorizes generation methods, and discusses countermeasures, emphasizing the imperative need for robust defences in safeguarding deep learning applications.\u003c/p\u003e\n\u003cp\u003eWieland Brendel et.al.[6] This study addresses the vulnerability of machine learning algorithms to imperceptible perturbations, emphasizing the risk in real-world applications. It introduces the Boundary Attack, a decision-based method applicable to black-box models, requiring minimal hyperparameter tuning and outperforming gradient-based attacks in tasks like ImageNet. The findings highlight the importance of studying model robustness and raise safety concerns for deployed machine learning systems.\u003c/p\u003e\n\u003cp\u003eGuofu Li et.al.[7] Adversarial machine learning investigates intentional data synthesis to mislead well-trained models, with a recent focus on imperceptible perturbations impacting deep learning accuracy. This paper offers a comprehensive overview of the field, covering foundations, attacking and defending strategies, and recent developments in deep learning\u0026apos;s susceptibility to adversarial attacks.\u003c/p\u003e\n\u003cp\u003eDou An et.al.[8] This paper tackles data integrity threats in smart grids through a Deep-Q-Network Detection (DQND) scheme, utilizing deep reinforcement learning for enhanced defense. DQND, overcoming dimensionality challenges, employs main and target networks for optimal defense strategy learning, displaying improved accuracy and speed in experimental evaluations on 9, 14, and 30 bus systems.\u003c/p\u003e\n\u003cp\u003eLingjuan Lyu et.al.[9] This article explores federated learning (FL) as a privacy-preserving alternative to centralized AI model training, highlighting the vulnerabilities of existing FL protocols to adversaries. The survey covers threat models, privacy attacks, defences, and poisoning attacks, providing insights into the evolving landscape of FL security over the past five years. It emphasizes the need for robust and privacy-ensuring FL systems in the face of diverse challenges.\u003c/p\u003e\n\u003cp\u003eAkhil Goel et.al.[10] This paper introduces a novel defence mechanism, a \u0026quot;defence layer,\u0026quot; to safeguard deep learning models from adversarial attacks, particularly in black-box and Gray-box settings. The parameter-free defence layer, applicable to various convolutional networks (CNNs), proves effective against attacks like FGSM, L_2, Elastic-Net, and Deep Fool. Experimental results on databases (MNIST, CIFAR-10, PaSC) and CNN architectures (VGG, ResNet, Dense Net) demonstrate the efficacy of the proposed defence layer without introducing computational overhead. For instance, on CIFAR-10, while an attack reduces ResNet-50 accuracy to 6.3%, the \u0026quot;defines layer\u0026quot; maintains an accuracy of 81.32%.\u003c/p\u003e"},{"header":"Methodology","content":"\u003cp\u003eThe data subsequent generations process encompasses the systematic acquisition and generation of data with the specific objective of facilitating research endeavours. The implementation of this task can be achieved through a diverse range of methodologies, including but not limited to the utilisation of surveys and experimentation techniques. The synthetic dataset employed in this research endeavour has been carefully crafted to closely emulate and reproduce the complex dynamics and characteristics observed in real-world scenarios. The numerical variables, denoted as\u0026nbsp;\u0026nbsp;to\u0026nbsp;\u0026nbsp;are obtained through the utilisation of diverse probability distributions to encompass an extensive range of data characteristics. The variables being examined in this study are referred to as variable1 and variable2. The data points were obtained through a sampling process from a standard normal distribution\u0026nbsp;, which is a widely employed statistical model for representing continuous features that demonstrate characteristic behaviour. A categorical variable consisting of two classes, denoted as 'A' and 'B', was randomly generated for the reason of introducing a categorical dimension into the research analysis. Variable3 is a crucial component of the research study being conducted. The identified variable demonstrates potential significance and relevance to the values that were sampled from a distribution that is uniform, specifically the\u0026nbsp;\u0026nbsp;distribution. This sampling technique was employed to introduce variability within a predetermined range. The term \"Variable4\" is commonly employed in the realm of research to refer to a particular variable that is under investigation within a specific study or experiment. It is of utmost significance to acknowledge that the simulation of a continuous variable was carried out through the utilisation of a normal distribution characterised by a mean value of zero and a standard deviation of one, conventionally represented as.The primary objective of the current investigation is to examine the phenomenon of simulating adversarial attacks. Adversarial attacks encompass intentional endeavours aimed at manipulating machine learning models through the introduction of meticulously crafted input data. In order to simulate adversarial threats, the researchers implemented deliberate attacks by introducing perturbations to a specific subset of the dataset. In the present inquiries, a comprehensive set of 20 samples was meticulously chosen through the implementation of a random sampling technique. The selection of these data points was conducted with the explicit objective of manipulating a range of numerical variables. The attack is effectively depicted and measured by employing the following mathematical expression:\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cimg src=\"data:image/png;base64,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\"\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eThe attacked value can be computed by adding the perturbation to the original value. The calculation of the attacked value involves the summation of the original value and the perturbation. The deliberate perturbation utilised within the scope of this research endeavour aims to replicate scenarios wherein an individual with malicious intent manipulates input data in order to deceive machine learning models. The statistical analyses performed in the present study were undertaken with the objective of investigating the associations and trends present within the gathered data. The current investigation utilised a range of statistical techniques to examine the data. Specifically, the emphasis of this evaluation is on the implementation of descriptive statistics. Descriptive statistics, as a fundamental branch of statistical analysis, endeavours to succinctly summarise and elucidate the primary attributes and features inherent within a given dataset. The calculation of essential statistical metrics, such as the mean \u0026nbsp;standard deviation and quartiles, is conducted for each numerical variable both before and after the adversarial attack. The t-test is a widely employed statistical test that is utilised to assess whether there exists a statistically significant distinction between the means of two groups or samples. A comparative analysis using a two-sample t-test was performed to assess the level of statistical significance in the differences observed in Variable1 between the 'A' and 'B' categories, both before and after the occurrence of the attack. The null hypothesis, commonly represented as , is formulated based on the underlying premise that there exists no statistically significant distinction between the variables under investigation. In contrast, the alternative hypothesis, commonly represented as , posits the presence of a substantial disparity between the variables under investigation. The employment of data visualisations is a widely adopted methodology across diverse domains, encompassing but not restricted to, data analysis, business intelligence, and scientific research. Boxplots are a commonly employed graphical tool for the purpose of visually examining and contrasting the distribution of numerical variables across different categories, both before and after the onset of an adversarial attack. Histograms provide researchers with valuable insights into the frequency distribution of variables prior to the start of an attack. The primary objective of these statistical examinations is to elucidate the influence of adversarial attacks on the basic characteristics of data and the outcomes produced by models. The aforementioned statement provides a fundamental basis for future discussions regarding the development of strategies aimed at preventing similar attacks and improving the robustness of models.\u003c/p\u003e"},{"header":"Result","content":"\u003cp\u003eThe present study\u0026apos;s findings provide valuable insights into the impact of adversarial attacks on machine learning models and the success rate of various defensive strategies. The study\u0026apos;s results can be categorised into two main components: descriptive stats and T-test Analysis. Descriptive statistics, as a fundamental branch of statistical analysis, encompasses the methodical gathering, arrangement, examination, and elucidation of data. The objective of this study is to provide a summary and description of the descriptive statistical data that were computed for each numerical variable, prior to as well as after the adversarial attack took place. Table 1 provides a succinct overview of the key statistical measures, including the mean\u0026nbsp;\u0026nbsp;deviation from the mean\u0026nbsp;\u0026nbsp;and quartiles. The tabular representation presented herein provides an extensive and informative representation of the fundamental characteristics inherent within the dataset under consideration.\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"612\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003e\u003cimg width=\"602\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp;\u003cimg width=\"605\" src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u0026nbsp;\u003cimg src=\"data:image/png;base64,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\" alt=\"image\"\u003e\u003c/p\u003e\n\u003cp\u003eThe correlation matrices that were observed offer a visual depiction of the alterations in the associations among parameters subsequent to the execution of the adversarial attack. Further investigation is imperative in order to elucidate the key variables that contribute to the previously mentioned changes and to comprehend their potential implications on the effectiveness of the model. The presented study\u0026apos;s findings provide significant benefits to the comprehension of the impacts of adversarial attacks on data distribution and the intricate relationship among variables. The aforementioned insights provide a solid basis for subsequent deliberations and inquiries pertaining to the robustness of models and the formulation of efficacious defence tactics\u003cstrong\u003e.\u003c/strong\u003e\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe focal point of the ongoing discourse is all about the subject matter of information for subsequent generations and attack simulation within the framework of code development. The specific aspect under consideration holds considerable significance within the realm of cybersecurity research and development. The objective of this discourse is to thoroughly examine the complexities associated with data generation as well as attack computer simulation methodologies, with a focus on their significance in evaluating the resilience and efficacy of safety precautions integrated within software code. The term \"data generation\" refers to the systematic procedure of generating synthetic data sets. In the current code implementation, a synthetic dataset is being created, consisting of five numerical variables that are labelled as Variable1 to Variable4, in addition to a categorical variable known as Category. Numerical variables are commonly derived from a diverse set of distributions order to accurately simulate and mimic real-world scenarios. The variable 'Category' consists of two discrete classes, denoted as 'A' and 'B', which have been randomly assigned. The process of simulating an adversarial attack involves intentionally introducing perturbations to a targeted subset within the dataset. For the purpose of this study, a sample size of twenty random samples was selected. The numerical variables present in each sample underwent a uniform increment by a constant value, specifically 5 units. The present scenario involves the execution of a simulated adversarial manipulation of input data, with the explicit objective of deceiving machine learning models. The code provided implements a t-test to assess the statistical significance of the differences in Variable1 values between categories A and B. This analysis conducted separately for the pre-attack and post-attack periods. The generation of boxplots and histograms serves the purpose of visually illustrating the distribution of each numerical variable across different categories prior to the onset of the attack. The preceding assertion provides significant observations regarding the attributes of the data being examined. The implementation of the defence mechanism involves the removal of samples that have been affected. Following the attack and subsequent defence, a series of new boxplots were generated to analyse the resulting changes in the distribution of the data. The presentation of summary statistics serves the purpose of providing a comprehensive overview of the dataset, encompassing key measures that encapsulate its central tendencies and variability. The current investigation involves the calculation and subsequent presentation of the correlation matrix before and after the attack event. Heatmaps have emerged as a highly valuable and indispensable tool within the domain of data visualization. They provide researchers with a powerful means to accurately represent and analyse the complex interdependencies that exist among different variables. By employing colour gradients to represent the intensity or magnitude of a particular variable, heatmaps offer a visually compelling way to uncover patterns, trends, and relationships that may otherwise remain hidden in raw data. This enables researchers to gain deeper insights and make informed decisions based on a comprehensive understanding of the data at hand. The examination of the effects of adversarial attacks on the interconnectedness of numerical variables holds significant significance in the pursuit of attaining a holistic comprehension of this phenomenon. The analysis of the findings involves leveraging the t-test results and visual depictions to acquire a deeper understanding of the impact of the adversarial attack on the data's attributes. The utilization of visual representations, such as boxplots and histograms, offers a comprehensive portrayal of the influence exerted by the attack on the distribution of numerical variables. The graphical displays employed in this study effectively communicate any alterations or deviations in the original distribution resulting from the attack. Moreover, the correlation matrices provide valuable insights into the changes observed in the associations among variables. Through a meticulous analysis of the correlation coefficients, it is possible to discern any discernible shifts or delays in the associations between various variables that have arisen as a consequence of the attack. The investigation delves into the ramifications of climate change on ecosystems at a global scale. The present inquiry delves into the pivotal role of genetic modification in the realm of agriculture. The examination of ethical considerations surrounding the utilization of animal testing in scientific research is a topic of great significance. The ethical implications of subjecting animals to experimentation for the advancement of scientific knowledge have been a subject of intense debate and scrutiny. This discourse revolves around the moral obligations and responsibilities that researchers and society at large bear towards animals, as well as the potential benefits and drawbacks associated with such practices. By delving into the ethical dimensions of animal testing, researchers aim to the investigation and exploration of the efficacy of adversarial attacks have garnered significant attention within the domains of machine learning and computer vision. The investigation into the exploitative capabilities of these attacks on deep learning models, with the intention of deceiving them and exerting influence, has been a subject of interest among researchers. Specifically, the impact of the adversarial attack on the statistically significant nature of Variable1 between categories A and B is deemed worthy of further examination and discussion. The investigation necessitates a comprehensive evaluation to determine the extent to which the attack successfully manipulated the data. Evaluation of defence Mechanisms: In this analysis, we will be examining the various defence mechanisms employed by individuals in order to cope with psychological stressors. Defence mechanisms are unconscious psychological strategies The evaluation of the simulated defence mechanism involves a thorough examination of the changes in data distribution that occur after the removal of the affected samples. Within the scope of this discussion, we shall undertake a comprehensive examination of the numerous constraints that have been identified within the given context, while also endeavouring to explore potential pathways for enhancement. In order to achieve a thorough comprehension of the potential obstacles that may impede the effectiveness or relevance of the impact analysis on data characteristics, it is essential to undertake a meticulous evaluation of these limitations. This examination can be facilitated by employing diverse statistical measures, as well as graphical representations like boxplots and histograms, which allow for the observation and interpretation of the attack's influence. The utilization of analytical tools facilitates the acquisition of valuable insights pertaining to alterations in the distribution and general statistics of the data. Summary statistics provide a succinct representation of key characteristics of a dataset, such as its central tendency, dispersion, and skewness. These measures, including the mean, median, standard deviation, and quartiles, offer valuable insights into the distribution and variability of the data. By examining these summary statistics, researchers can gain a comprehensive understanding of the dataset's overall pattern and characteristics. Through a comparative analysis of the summary statistics prior to and subsequent to the occurrence of the attack, it is possible to discern any noteworthy modifications in these quantitative measures. In the context of data analysis, it is worth noting that a shift in the mean or an increase in the standard deviation can potentially signify a disturbance in the underlying pattern of the data. Boxplots have been widely recognized as a valuable tool in data visualization, offering researchers a comprehensive understanding of data distribution and facilitating the identification of any potential outliers. Through a careful analysis of the boxplots both prior to and subsequent to the occurrence of the attack, we are able to evaluate any potential alterations in the dispersion, asymmetry, or existence of outliers. The current analysis focuses on investigating any noteworthy deviations within the dataset. Specifically, our objective is to explore the alterations in the relationship between variables by examining the changes that have been noticed in the correlation matrices. The careful consideration of the implications pertaining to machine learning algorithms that rely on these relationships is of utmost importance. Real-world applications refer to the practical utilization of a concept, theory, or technology in various domains outside of academic or theoretical settings. These applications the current investigation establishes a fundamental correlation between the acquired results and their pragmatic ramifications. The understanding and mitigation of adversarial attacks in the field of machine learning are of utmost importance.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn this study has explored the complex dynamics associated with adversarial attacks targeting machine learning models. Specifically, it has examined the effects of these attacks on the characteristics of the data and the subsequent countermeasures employed for defence. The present study provides a comprehensive analysis of the key findings and their corresponding implications. The following section presents a concise summary of these findings and their potential significance in the broader research context. The effects of adversarial attacks were examined in this study. The key findings indicate that these attacks can have significant impacts on the performance and robustness of machine learning models. Specifically, it was observed that adversarial attacks can lead to a decrease in accuracy and the findings of the conducted simulated adversarial attack, as indicated by the statistical analysis using t-test, revealed a noticeable and statistically significant effect on Variable1 when comparing categories, A and B. The aforementioned observation highlights the susceptibility of machine learning models to deliberate manipulations of data. The phenomenon under investigation pertains to shifts in data distribution. The examination of visual representations, specifically boxplots and histograms, provided valuable insights into the notable alterations in the distribution patterns of numerical variables both prior to and subsequent to the occurrence of the adversarial attack. The aforementioned observation underscores the possibility of feature representations being distorted, which in turn can significantly impact the decision-making process of a model. The present study aims to conduct a comprehensive assessment of defence mechanisms. Defense mechanisms are psychological strategies employed by individuals to cope with various internal and external stressors. Understanding these mechanisms can provide the observed defence mechanism, which involved the removal of affected samples, exhibited a certain level of efficacy in the restoration of data integrity. Further investigation and refinement of defence strategies are necessary in order to improve their resilience and effectiveness. The present study aims to investigate potential alterations in the correlation matrix. The examination of correlation matrices has unveiled notable alterations in the associations among numerical variables subsequent to the implementation of attack and defence strategies. Comprehending these modifications is of utmost importance in evaluating the dependability of model forecasts when confronted with adversarial manipulations. The implications of the aforementioned findings are of significant importance and warrant further investigation. The results suggest potential ramifications The investigation of real-world model resilience is a topic of great significance in the field of research. The examination of how models perform and adapt in real-world scenarios is crucial for understanding the present study highlights the practical implications of adversarial attacks, thereby emphasizing the crucial need for robust defence mechanisms in order to protect machine learning models in real-world scenarios. Data quality assurance is a crucial aspect of any research study or data analysis process. It involves ensuring the accuracy, reliability, and consistency of the data being used. By implementing various quality, the examination of adversarial attacks and their effects on data distribution sheds light on the necessity of continuous endeavours in guaranteeing the quality and integrity of data, especially in situations where machine learning models hold significant importance. The user's text is concise and lacks specific information regarding the contributions made to a particular field. Further elaboration is required to provide a comprehensive understanding of the contributions made by individuals or groups. The present study makes a valuable contribution to the ever-changing field of machine learning security by conducting a comprehensive investigation into the ramifications of adversarial attacks on both data and model outputs. The findings presented in this study provide valuable insights for researchers, practitioners, and policymakers who are interested in enhancing the resilience of machine learning systems against deliberate manipulations. In summary, this research study provides a foundational basis for future investigations into the comprehension of adversarial threats and the formulation of advanced defence tactics to enhance the robustness of machine learning models in dynamic and arduous settings.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eImplemented the adversarial attack simulation, generated example data, and conducted statistical tests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eRosenberg, I., Shabtai, A., Elovici, Y., \u0026amp; Rokach, L. (2021). Adversarial machine learning attacks and defense methods in the cyber security domain. ACM Computing Surveys (CSUR), \u003cem\u003e54\u003c/em\u003e(5), 1\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXue, M., Yuan, C., Wu, H., Zhang, Y., \u0026amp; Liu, W. (2020). Machine learning security: Threats, countermeasures, and evaluations. IEEE Access, 8, 74720\u0026ndash;74742.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Q., Li, P., Zhao, W., Cai, W., Yu, S., \u0026amp; Leung, V. C. (2018). A survey on security threats and defensive techniques of machine learning: A data driven view. IEEE access, 6, 12103\u0026ndash;12117.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChakraborty, A., Alam, M., Dey, V., Chattopadhyay, A., \u0026amp; Mukhopadhyay, D. (2018). Adversarial attacks and defences: A survey. arXiv preprint arXiv:1810.00069.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan, X., He, P., Zhu, Q., \u0026amp; Li, X. (2019). Adversarial examples: Attacks and defenses for deep learning. IEEE transactions on neural networks and learning systems, 30(9), 2805\u0026ndash;2824.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrendel, W., Rauber, J., \u0026amp; Bethge, M. (2017). Decision-based adversarial attacks: Reliable attacks against black-box machine learning models. arXiv preprint arXiv:1712.04248.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, G., Zhu, P., Li, J., Yang, Z., Cao, N., \u0026amp; Chen, Z. (2018). Security matters: A survey on adversarial machine learning. arXiv preprint arXiv:1810.07339.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAn, D., Yang, Q., Liu, W., \u0026amp; Zhang, Y. (2019). Defending against data integrity attacks in smart grid: A deep reinforcement learning-based approach. IEEE Access, 7, 110835\u0026ndash;110845.\u003c/span\u003e\u003c/li\u003e \u003cli\u003eLyu, L., Yu, H., Ma, X., Chen, C., Sun, L., Zhao, J., ... \u0026amp; Philip, S. Y. (2022). Privacy and robustness in federated learning: Attacks and defenses. IEEE transactions on neural networks and learning systems.\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGoel, A., Agarwal, A., Vatsa, M., Singh, R., \u0026amp; Ratha, N. K. (2020). DNDNet: Reconfiguring CNN for adversarial robustness. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 22\u0026ndash;23).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Adversarial Attacks, Artificial Intelligence Security, Cryptography, Secure Federated Learning, Model Forensics","lastPublishedDoi":"10.21203/rs.3.rs-3871878/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3871878/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe primary objective of the research is to examine the resilience of models based on machine learning when confronted with adversarial threats. The focus will be on developing effective strategies to safeguard the integrity of these models. This study employs a carefully designed framework to undertake the task of simulating adversarial attacks on machine learning models. The central objective of the research centres on the intentional manipulation of numerical variables in order to accurately simulate real-world threats. The investigation of statistical analyses, such as t-tests and comprehensive data visualisations, enables the evaluation of the impact exerted by adversarial attacks on both data distributions and model outputs. The primary objective of the current investigation is to assess the effectiveness of a mimicked defence mechanism in the domain of machine learning model restoration, with the ultimate goal of preserving model integrity. The results gathered by our study emphasise the significant significance of adversarial threats, thereby underscoring the need for a comprehensive analysis of robust defence strategies. The analysis of correlation matrices, both before and after adversarial attacks, provides valuable insights into the changes in the fundamental structure induced by these interventions. The current investigation offers valuable insights into the preservation of the integrity of machine learning models and suggests possible ways to enhance their resilience against ever-changing adversarial tactics.\u003c/p\u003e","manuscriptTitle":"Defending the Integrity of Machine Learning Models in the Face of Adversarial Threats","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-01-19 08:52:16","doi":"10.21203/rs.3.rs-3871878/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"daf199e4-a2d4-4a3d-969d-10b8b0e4dd05","owner":[],"postedDate":"January 19th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2024-02-12T07:59:56+00:00","versionOfRecord":[],"versionCreatedAt":"2024-01-19 08:52:16","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3871878","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3871878","identity":"rs-3871878","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}