Sravanti Addepalli

Papers

Total Citations

H-Index

About

Sravanti Addepalli’s research lies at the intersection of adversarial machine learning and robust deep neural network design, with a focus on bridging the gap between theoretical defenses and real-world safety-critical applications. Her most-cited work, “Boosting Adversarial Robustness using Feature Level Stochastic Smoothing” (2021, 4 citations), introduces a novel defense mechanism that applies stochastic smoothing at the feature level rather than the input space, significantly enhancing model resilience against adversarial perturbations. This contribution is particularly impactful for domains like robotics and autonomous navigation, where even small vulnerabilities can lead to catastrophic failures. Addepalli’s approach addresses a key limitation in existing adversarial training methods by improving robust accuracy without sacrificing clean performance. Her work has been recognized for its practical relevance, earning citations from researchers advancing both theoretical foundations and deployment-ready defenses. By tackling the persistent robustness gap in state-of-the-art models, Addepalli is helping pave the way for more trustworthy AI systems in high-stakes environments. Her research continues to influence the development of scalable, efficient adversarial defenses that are critical for the next generation of autonomous technologies.