Optimization
Nonparametric Evaluation of Noisy ICA Solutions
Independent Component Analysis (ICA) was introduced in the 1980's as a model
Exogenous Matching: Learning Good Proposals for Tractable Counterfactual Estimation
"why," is crucial for attribution and explanation,
Why Transformers Need Adam: A Hessian Perspective
CNNs, MLPs, and quadratic problems, and find that SGD can perform on par with Adam on problems without block heterogeneity, but performs worse than Adam when the heterogeneity exists.
Sharpness-A ware Minimization Activates the Interactive Teaching's Understanding and Optimization
However, the underlying optimization principles and convergence of interactive teaching lack theoretical analysis, and in this regard co-teaching serves as a notable prototype.
On Causal Discovery in the Presence of Deterministic Relations
In this paper, we find, supported by both theoretical analysis and empirical evidence, that score-based methods with exact search can naturally address the issues of deterministic relations under rather mild assumptions. Nonetheless, exact score-based methods can be computationally expensive.
Minimizing UCB: a Better Local Search Strategy in Local Bayesian Optimization
Local Bayesian optimization is a promising practical approach to solve high dimensional black-box function optimization problem. Among them is the approximated gradient class of methods, which implements a strategy similar to gradient descent. These methods have achieved good experimental results and theoretical guarantees.
Self-Guiding Exploration for Combinatorial Problems
They utilize prompting techniques such as Exploration-of-Thought, Decomposition, and Refinement to effectively navigate and solve intricate tasks.
Block Sparse Bayesian Learning: A Diversified Scheme
This paper introduces a novel prior called Diversified Block Sparse Prior to characterize the widespread block sparsity phenomenon in real-world data.
