Title:  Probabilistic Reasoning in Complex Systems: Algorithms and Applications

Abstract:

Statistical machine learning approaches to scientific applications are complicated by the high-dimensional, continuous, and nonlinear interactions that typically arise. In this talk I will discuss several general-purpose algorithms for statistical inference in probabilistic graphical models with these properties. I will introduce Diverse Particle Max-Product (D-PMP), a particle-based extension of max-product belief propagation (BP) for maximum a posteriori (MAP) inference and will show how D-PMP isolates multiple distinct, locally-optimal, configurations through a diverse particle selection process. I will further demonstrate how D-PMP can be flexibly adapted to problems of articulated human pose estimation in images and video, as well as protein structure prediction from low-resolution experimental data. In the second half of the talk I will present robust and efficient algorithms for sequential decision making in information gathering systems. These methods sequentially select actions which maximally reduce posterior uncertainty at each stage. Using gene regulatory network inference as an example, I will demonstrate how these general-purpose algorithms meet or exceed the performance of domain-specific approaches for Bayesian experimental design.

Bio: 

Jason Pacheco is a postdoctoral associate at MIT in the Computer Science and Artificial Intelligence Laboratory (CSAIL) with John Fisher III. Prior to joining MIT Jason completed his graduate work at Brown University with Erik Sudderth. Jason’s research interests are in statistical machine learning, probabilistic graphical models, approximate inference algorithms, and information-theoretic decision making.