Learning Graphical Models
An Analysis of Elo Rating Systems via Markov Chains
This amount depends on the previously estimated difference in skills between i and j .
Improving Generalization in Federated Learning with Model-Data Mutual Information Regularization: A Posterior Inference Approach
Theoretical analysis provides a generalization bound for FL w.r.t. the
A Local Method for Satisfying Interventional Fairness with Partially Known Causal Graphs
To exploit the PDAGs for achieving interventional fairness, previous methods have been built on variable selection or causal effect identification, but limited to reduced prediction accuracy or strong assumptions. In this paper, we propose a general min-max optimization framework that can achieve interventional fairness with promising prediction accuracy and can be extended to maximally oriented PDAGs (MPDAGs) with added background knowledge.
The Bayesian sampling in a canonical recurrent circuit with a diversity of inhibitory interneurons
Accumulating evidence suggests stochastic cortical circuits can perform sampling-based Bayesian inference to compute the latent stimulus posterior. Canonical cortical circuits consist of excitatory (E) neurons and types of inhibitory (I) in-terneurons. Nevertheless, nearly no sampling neural circuit models consider the diversity of interneurons, and thus how interneurons contribute to sampling remains poorly understood.
Unrolled denoising networks provably learn to perform optimal Bayesian inference
Unrolled AMP architectures and variants thereof were originally proposed and empirically investigated by Borgerding et al. [
Beating Adversarial Low-Rank MDPs with Unknown Transition and Bandit Feedback
Previous work has investigated this problem under either full-information loss feedback with unknown transitions (Zhao et al., 2024), or bandit
