About

Vishnu Dasu is a highly motivated and skilled researcher with a strong publication record, currently working as a Graduate Teaching Assistant at Penn State University. He holds a Master of Science in Computer Science from Penn State University and a Bachelor of Technology in Computer Science from Manipal Institute of Technology. Vishnu's research interests lie at the confluence of Machine Learning and Cybersecurity. He has worked as a Researcher at TCS in the Cybersecurity and Privacy division on anomaly and insider threat detection using ML techniques like graph neural networks and natural language processing. He has also worked under Dr. Sumanta Sarkar on privacy-preserving federated learning, where he developed a single round, fault-tolerant secure aggregation protocol for federated learning. Additionally, he has explored secure aggregation and functional encryption protocols to determine their feasibility in federated learning scenarios. Vishnu has received the prestigious "TCS Citation Award" three times for his outstanding contribution to the organization and for publishing high-quality research papers. As a Research Intern at TCS, Vishnu worked on preventing adversarial attacks on neural networks. He looked into the feasibility of using denoising autoencoders in removing adversarial noise added to ResNet-based architectures. His approach using denoising autoencoders was 86% effective in removing adversarial noise. Vishnu's technical skills include Research Scientist, ML, NLP, Neural Networks, Software Engineer, and CI/CD. He is open to collaborations and would love to interact with like-minded researchers.