My research broadly spans machine learning, deep learning and signal processing for applications in computer vision, healthcare, graph analysis and multimodal learning. I am with the Machine Intelligence group at Lawrence Livermore National Laboratories. As part of my research, I focus on unsupervised/representation learning, OOD generalization, uncertainty quantification, ML robustness/safety, generative models, transfer learning, few-shot learning and inverse problems.
If you are looking for postdoc opportunities in ML/AI (unsupervised learning, safe ML, knowledge-aware AI and scientific machine learning), reach out to me!
Some exciting projects that I have been part of recently
AI-Powered Cognitive Simulations to Accelerate Scientific Discovery (High-Energy physics, Material Science, Computational Biology, Epidemiology)
Uncertainty Quantification in Deep Neural Networks to Advance Black-Box Optimization and to Improve ML Safety
Knowledge-Integrated Learning to Build Data-Efficient and Well Grounded AI Solutions for Biology and Situational Awareness
Diagnostic AI Tools for Traumatic Brain Injury, Cancer and Patient Health Monitoring (Human Connectomes, Multimodal Imaging Data, Electronic Health Records)
Advanced Algorithms for Robust ML, Domain Generalization, Test-Time Adaptation, Few shot learning and Transfer Learning
Explainability Tools to Support Human-Centric Analysis of Deep Neural Networks
Solving Inverse Optimization for Ill-posed Image Recovery, Audio Restoration and High-Dimensional Parameter Space Exploration
High-dimensional Sampling for Experiment Design (Additive Manufacturing) and Analyzing Generalization of ML Models