Polyak--Ruppert
PulseAugur coverage of Polyak--Ruppert — every cluster mentioning Polyak--Ruppert across labs, papers, and developer communities, ranked by signal.
1 day(s) with sentiment data
-
New SGD method enables online quantile estimation with theoretical guarantees
This paper introduces a novel smoothed stochastic gradient descent (SGD) algorithm for online quantile estimation. The method ensures estimates remain monotone with respect to the quantile level throughout the streaming…
-
Two arXiv papers explore finite-iteration theory and online inference for temporal-difference learning
Two new arXiv papers delve into the theoretical underpinnings of temporal-difference (TD) learning methods, focusing on their finite-iteration behavior and online statistical inference. The first paper by Ege Can Kaya a…
-
New arXiv papers explore advanced quantile regression with privacy and applications
Two new research papers on arXiv introduce advanced quantile regression techniques. The first paper details pairwise quantile regression, establishing theoretical guarantees and demonstrating its application in facial r…
-
New TD(0) algorithm achieves robust and fast convergence with single stepsize
Researchers have developed a new method for linear TD(0) algorithms that uses a single stepsize schedule, eliminating the need for prior knowledge of curvature parameters. This approach provides high-probability guarant…
-
New Theory: SA-Adam Adaptivity Asymptotically Invisible
Researchers have published a paper detailing a theoretical analysis of adaptive optimization algorithms, specifically focusing on SA-Adam with momentum and non-convergent adaptive preconditioning. The study proves a non…
-
New Q-learning method achieves n^{-1/4} Gaussian approximation bound
Researchers have developed a new method for approximating Gaussian distributions in entropy-regularized Q-learning with function approximation. The study establishes convergence rates for averaged iterates generated by …
-
Researchers develop novel bootstrap for SGD confidence sets
Researchers have developed a novel method for constructing confidence sets in Stochastic Gradient Descent (SGD) algorithms. This new approach utilizes the multiplier bootstrap procedure and establishes its non-asymptoti…
-
New research identifies stabilization threshold for dynamic preconditioning in online inference
Researchers have identified a critical stabilization threshold for dynamic preconditioning in gradient descent methods. This threshold determines when the Polyak-Ruppert averaging technique, fundamental for online infer…