Aditi Krishnapriyan (UC Berkeley): Schmidt AI in Science Speaker Series

Title: Learning physical dynamics with generative machine learning

Abstract: Recent advances in large-scale scientific datasets are creating new opportunities for machine learning (ML) methods to more effectively capture scientific phenomena with greater accuracy and reach. In this talk, I will discuss how these advances are both shifting ML design paradigms and enabling new scientific inquiries. This includes investigations into understanding if neural networks, and specifically Transformers, can autonomously discover fundamental physical relationships from data, and demonstrating how more flexible machine learning modeling design choices enable capturing physical dynamics across multiple scales. I will also explore how generative modeling approaches grounded in statistical mechanics can be applied to accelerate the sampling of transition pathways, and as a framework to align and bridge the gap between numerically simulated data and experimental observations.

Bio: Aditi Krishnapriyan, Assistant Professor, University of California, Berkeley. I  am interested in developing methods in machine learning that are driven by the distinct challenges and opportunities in the natural sciences, with particular interest in physics-inspired machine learning methods. Some areas of exploration include general learning strategies exploring the relevance of physical inductive biases and ML models for scientific problems, the advantages that ML can bring to classical physics-based numerical solvers (such as through end-to-end differentiable frameworks and implicit layers), and better learning strategies for distribution shifts in the physical sciences. These methods are informed by and grounded in applications in atomistic and continuum problems, including fluid mechanics, molecular dynamics, and other related areas. This work also includes interfacing with other fields including numerical analysis, dynamical systems theory, quantum mechanics, computational geometry, optimization, and category theory.