Grainger College of Engineering, All Events

Abstract: Generative models, such as large language models and diffusion models, have tremendously increased the scope of problems that AI can address. As such, there is a significant trend toward incorporating generative AI to automate tasks across computing and more broadly, from controlling robotics systems, to software generation and testing, to searching over scientific knowledge. However, there remains a significant formalization gap between the domain knowledge, theories, and logical and semantic constraints that are vital to applications, and the statistical patterns over natural data represented by large generative models. In this talk, I will demonstrate how we can systematically bridge this formalization gap towards more trustworthy AI. First, drawing from examples and applications in my research, I will show how we can utilize tractable intermediate representations of probability distributions to bridge between formal language and generative models at scale. Then, I will discuss how these practical methods are underpinned by advances in the mathematical and computational foundations underlying these tractable representations of probability distributions.

Bio: Benjie Wang is a Postdoctoral Researcher in the Statistical and Relational Artificial Intelligence (StarAI) Lab at UCLA. Previously, he was a research fellow at the Simons Institute for the Theory of Computing at UC Berkeley, and completed his PhD at the University of Oxford advised by Marta Kwiatkowska. His research interests lie broadly in artificial intelligence, spanning deep generative models, probabilistic machine learning, natural language processing, and computational mathematics. His work on these topics has been published in top venues including NeurIPS, ICML, ICLR, JMLR, and EMNLP, and has been recognized by selective oral presentations and a best workshop paper award.