CSML Seminar Series
- Mon, Apr 19, 2021, 4:30 pmWhile exciting progress has been made in understanding the global convergence of vanilla gradient methods for solving challenging nonconvex problems in statistical estimation and machine learning, their computational efficacy is still far from satisfactory for ill-posed or ill-conditioned problems. In this talk, we discuss how the trick of preconditioning further boosts the convergence speed with minimal computation overheads through two examples: low-rank matrix estimation in statistical learning and policy optimization in entropy-regularized reinforcement learning.
- Fri, Jan 15, 2021, 4:30 pmIn this talk, we offer an entirely “white box’’ interpretation of deep (convolutional) networks from the perspective of data compression. In particular, we show how modern deep architectures, linear (convolution) operators and nonlinear activations, and parameters of each layer can be derived from the principle of rate reduction (and invariance).
- Fri, Mar 6, 2020, 12:00 pm
Psychology has traditionally been a laboratory discipline, focused on small-scale experiments conducted in person. However, recent technological innovations have made it possible to collect far more data from far more people than ever before. In this talk, I will explore some of the consequences of being able to conduct psychological research at a larger scale, highlighting some of the tools that we have developed for doing so. In particular, I will focus on three recent projects exploring aspects of human decision-making.
- Mon, Feb 24, 2020, 4:00 pm
This talk considers the preference modeling problem and addresses the fact that pairwise comparison data often reflects irrational choice, e.g. intransitivity. Our key observation is that two items compared in isolation from other items may be compared based on only a salient subset of features. Formalizing this idea, I will introduce our proposal for a “salient feature preference model” and discuss sample complexity results for learning the parameters of our model and the underlying ranking with maximum likelihood estimation.
- Mon, Feb 10, 2020, 4:00 pmSemidefinite programming (SDP) is a powerful framework from convex optimization that has striking potential for data science applications. This talk develops new provably correct algorithms for solving large SDP problems by economizing on both the storage and the arithmetic costs. We present two methods: one based on sketching, and the other on complementarity. Numerical evidence shows that these methods are effective for a range of applications, including relaxations of MaxCut, abstract phase retrieval, and quadratic assignment, and can handle SDP instances where the matrix variable has over 10^13 entries.
- Fri, Feb 21, 2020, 12:15 pmImportant insights into redistricting can be gained by formulating and analyzing the problem with a Markov Chain Monte Carlo framework that utilizes optimization heuristics to inform transition proposals.
- Fri, Feb 7, 2020, 12:00 pmWe review targeted minimum loss estimation (TMLE), which provides a general template for the construction of asymptotically efficient plug-in estimators of a target estimand for infinite dimensional models.
- Mon, Feb 10, 2020, 4:30 pmThe human body responds to neurocognitive stress in multiple ways through its autonomic nervous system. Changes in skin conductance measurements indicate sudomotor nerve activity, and could be used in inferring the underlying autonomic nervous system stimulation. We model skin conductance measurements using a state-space model with sparse impulsive events as inputs to the model.
- Fri, Nov 8, 2019, 12:30 pmDiscrete structure recovery is an important topic in modern high-dimensional inference. Examples of discrete structure include clustering labels, ranks of players, and signs of variables in a regression model.
- Tue, Nov 12, 2019, 4:30 pm
In many real-world problems learners cannot directly observe their own rewards but can still infer whether some particular action is successful. How should a learner take actions to balance its need of information while maximizing their reward in this setting? Partial monitoring is a framework introduced a few decades ago to model learning situations of this kind. The simplest version of partial monitoring generalizes learning with full-, bandit-, or even graph-information based feedback.