Samyak Jain

International Institute of Information Technology, Indian Institute of Science Bangalore

Papers

Total Citations

H-Index

About

Samyak Jain’s research lies at the intersection of visual cognition and trustworthy AI, focusing on computational models of visual attention and adversarial robustness in deep neural networks. In his influential work on saliency prediction, Jain critically analyzed deep architectures to streamline models that mimic human visual attention, advancing efforts to equip machines with human-like perceptual abilities. His paper “Tidying Deep Saliency Prediction Architectures” has garnered 4 citations, reflecting its foundational role in refining data-driven saliency approaches. Addressing the critical challenge of security in AI, Jain’s work “Boosting Adversarial Robustness using Feature Level Stochastic Smoothing” (4 citations) introduces a novel defense mechanism that enhances robustness against adversarial attacks—a vital step for deploying neural networks in high-stakes domains like robotics and autonomous navigation. By tackling both the cognitive and safety dimensions of deep learning, Jain demonstrates a commitment to building not only more capable but also more reliable AI systems. His contributions are particularly relevant for researchers exploring the intersection of human vision and machine learning, as well as those working on secure AI for real-world applications.