ROBUSTNESS AND ADVERSARIAL ATTACKS ON GENERATIVE MODELS

Yeshwanth Vasa

doi:10.36676/jrps.v12.i3.1537

Authors

Yeshwanth Vasa Independent Researcher

DOI:

https://doi.org/10.36676/jrps.v12.i3.1537

Keywords:

Stability, Adversarial Perturbation, Model Generation, Generative Adversarial Networks, Variational Autoencoder, Adversarial Optimization, Anti-Adversarial Measures

Abstract

Since generative models rely on providing input data samples, it is essential to have a robust generative model capable of standing against adversarial attacks that can tamper with the model's output. This paper employs empirical analysis to examine the weaknesses of critical generative models like GANs and VAEs and additionally discovers the defense schemes. In a controlled environment created by accurately modeled adversarial trial data sets and time-sensitive analyses, we test and compare various confirmed adversarial training methods and defenses, such as implicit generative modeling and probabilistic adversarial robustness. Our results emphasize the difficulty of gaining complete robustness and suggest a way to deal with such attacks while preserving the model's accuracy. The analysis also reveals gaps in existing techniques, opening up possibilities for future research to improve the protection of generative models. This work will be valuable for the machine learning community in the future, as it contributes to discussing adversarial robustness and offers insights for researchers and practitioners.

References

Carlini, N., Athalye, A., Papernot, N., Brendel, W., Rauber, J., Tsipras, D., ... & Kurakin, A. (2019). On evaluating adversarial robustness. arXiv preprint arXiv:1902.06705. https://arxiv.org/pdf/1902.06705

Katikireddi, P. M., Singirikonda, P., & Vasa, Y. (2021). Revolutionizing DEVOPS with Quantum Computing: Accelerating CI/CD pipelines through Advanced Computational Techniques. Innovative Research Thoughts, 7(2), 97–103. https://doi.org/10.36676/irt.v7.i2.1482 DOI: https://doi.org/10.36676/irt.v7.i2.1482

Vasa, Y., Jaini, S., & Singirikonda, P. (2021). Design Scalable Data Pipelines For Ai Applications. NVEO - Natural Volatiles & Essential Oils, 8(1), 215–221. https://doi.org/https://doi.org/10.53555/nveo.v8i1.5772 DOI: https://doi.org/10.53555/nveo.v8i1.5772

Singirikonda, P., Jaini, S., & Vasa, Y. (2021). Develop Solutions To Detect And Mitigate Data Quality Issues In ML Models. NVEO - Natural Volatiles & Essential Oils, 8(4), 16968–16973. https://doi.org/https://doi.org/10.53555/nveo.v8i4.5771 DOI: https://doi.org/10.53555/nveo.v8i4.5771

Vasa, Y. (2021). Develop Explainable AI (XAI) Solutions For Data Engineers. NVEO - Natural Volatiles & Essential Oils, 8(3), 425–432. https://doi.org/https://doi.org/10.53555/nveo.v8i3.5769 DOI: https://doi.org/10.53555/nveo.v8i3.5769

Jangampeta, S., Mallreddy, S. R., & Padamati, J. R. (2021). Data Security: Safeguarding the Digital Lifeline in an Era of Growing Threats. International Journal for Innovative Engineering and Management Research, 10(4), 630-632.

Sukender Reddy Mallreddy(2020).Cloud Data Security: Identifying Challenges and Implementing Solutions.JournalforEducators,TeachersandTrainers,Vol.11(1).96 -102.

Nunnaguppala, L. S. C. , Sayyaparaju, K. K., & Padamati, J. R.. (2021). "Securing The Cloud: Automating Threat Detection with SIEM, Artificial Intelligence & Machine Learning", International Journal For Advanced Research In Science & Technology, Vol 11 No 3, 385-392

Padamati, J., Nunnaguppala, L., & Sayyaparaju, K. . (2021). "Evolving Beyond Patching: A Framework for Continuous Vulnerability Management", Journal for Educators, Teachers and Trainers, 12(2), 185-193.

Nunnaguppala, L. S. C. . (2021). "Leveraging AI In Cloud SIEM And SOAR: Real-World Applications For Enhancing SOC And IRT Effectiveness", International Journal for Innovative Engineering and Management Research,10(08), 376-393