Systems | Information | Learning | Optimization
 

SILO: Data Science Institute Talks

Bio Kyle Cranmer is a professor in the Physics Department with affiliate appointments in Computer Sciences and Statistics. He is also the David R. Anderson Director of the University of Wisconsin-Madison’s Data Science Institute (DSI). Professor Cranmer obtained his Ph.D. in Physics from the University of Wisconsin-Madison in 2005 and …

SILO: Understanding and Leveraging Adaptive Algorithms’ Sensitivity to Change-of-Basis

Abstract Adaptive gradient methods—such as Adagrad, Adam, and their variants—have found widespread use in machine learning, signal processing, and many other settings. However many algorithms in this family are not rotationally equivariant: in this talk we examine how a simple change-of-basis in either parameter space or data space can drastically …

SILO: Optimization over Trained Neural Networks: Going Large with Gradient-Based Algorithms

Abstract When optimizing a nonlinear objective, one can employ a neural network as a surrogate for the nonlinear function. However, the resulting optimization model can be time-consuming to solve globally with exact methods. As a result, local search that exploits the neural-network structure has been employed to find good solutions …

SILO: Finite‑Time Bounds for Robust Reinforcement Learning with Linear Function Approximation

Abstract Robust reinforcement learning (RL) focuses on designing optimal policies from data for MDPs with model uncertainties. Existing convergence guarantees for robust RL are either limited to tabular settings or use restrictive assumptions in the function approximation setting. We will present an RL algorithm for learning the optimal policy from …

SILO: *Canceled*

Abstract From Entropy to Epiplexity: Rethinking Information for Computationally Bounded Intelligence. Can we learn more from data than existed in the generating process itself? Can new and useful information be constructed from merely applying deterministic transformations to existing data? Can the learnable content in data be evaluated without considering a …