no code implementations • 17 Jul 2023 • Ziwei Su, Raghu Pasupathy, Yingchieh Yeh, Peter W. Glynn
The message from extensive numerical experimentation is that in settings where a functional CLT on the point estimator is in effect, using \emph{large overlapping batches} alongside OB-x critical values yields confidence intervals that are often of significantly higher quality than those obtained from more generic methods like subsampling or the bootstrap.
no code implementations • 13 Dec 2022 • Hongseok Namkoong, Yuanzhe Ma, Peter W. Glynn
The performance of decision policies and prediction models often deteriorates when applied to environments different from the ones seen during training.
no code implementations • 17 Oct 2022 • Li Xia, Peter W. Glynn
CVaR (Conditional Value at Risk) is a risk metric widely used in finance.
no code implementations • 11 Oct 2022 • Lin Fan, Peter W. Glynn
The tail characterizations there are associated with atypical bandit behavior on trajectories where the optimal arm mean is under-estimated, leading to mis-identification of the optimal arm and large regret.
no code implementations • NeurIPS 2021 • Carson Kent, Jiajin Li, Jose Blanchet, Peter W. Glynn
We propose a novel Frank-Wolfe (FW) procedure for the optimization of infinite-dimensional functionals of probability measures - a task which arises naturally in a wide range of areas including statistical learning (e. g. variational inference) and artificial intelligence (e. g. generative adversarial networks).
no code implementations • 28 Sep 2021 • Lin Fan, Peter W. Glynn
It is well known that designs that are optimal over certain exponential families can achieve expected regret that grows logarithmically in the number of arm plays, at a rate governed by the Lai-Robbins lower bound.
no code implementations • 6 Jul 2021 • Zhengyuan Zhou, Panayotis Mertikopoulos, Nicholas Bambos, Peter W. Glynn, Yinyu Ye
One of the most widely used methods for solving large-scale stochastic optimization problems is distributed asynchronous stochastic gradient descent (DASGD), a family of algorithms that result from parallelizing stochastic gradient descent on distributed computing architectures (possibly) asychronously.
no code implementations • 19 May 2021 • Lin Fan, Peter W. Glynn
In the regime where the gaps between arm means scale as $1/\sqrt{n}$ with the time horizon $n$, we show that the dynamics of Thompson sampling evolve according to discrete versions of SDEs and random ODEs.
no code implementations • 12 Mar 2021 • Jacques A. de Chalendar, Peter W. Glynn
Weather forecast information will very likely find increasing application in the control of future energy systems.
no code implementations • 14 Apr 2020 • Yanjun Han, Zhengqing Zhou, Zhengyuan Zhou, Jose Blanchet, Peter W. Glynn, Yinyu Ye
We study the sequential batch learning problem in linear contextual bandits with finite action sets, where the decision maker is constrained to split incoming individuals into (at most) a fixed number of batches and can only observe outcomes for the individuals within a batch at the batch's end.
no code implementations • NeurIPS 2018 • Zhengyuan Zhou, Panayotis Mertikopoulos, Susan Athey, Nicholas Bambos, Peter W. Glynn, Yinyu Ye
We consider a game-theoretical multi-agent learning problem where the feedback information can be lost during the learning process and rewards are given by a broad class of games known as variationally stable games.
no code implementations • 7 Sep 2018 • Lin Fan, Peter W. Glynn, Markus Pelger
We propose novel methods for change-point testing for nonparametric estimators of expected shortfall and related risk measures in weakly dependent time series.
no code implementations • NeurIPS 2017 • Zhengyuan Zhou, Panayotis Mertikopoulos, Nicholas Bambos, Peter W. Glynn, Claire Tomlin
We consider a model of game-theoretic learning based on online mirror descent (OMD) with asynchronous and delayed feedback information.
no code implementations • NeurIPS 2017 • Zhengyuan Zhou, Panayotis Mertikopoulos, Nicholas Bambos, Stephen Boyd, Peter W. Glynn
In this paper, we examine a class of non-convex stochastic optimization problems which we call variationally coherent, and which properly includes pseudo-/quasiconvex and star-convex optimization problems.
no code implementations • 26 Dec 2013 • Mohammad Mousavi, Peter W. Glynn
We present a fully nonparametric method to estimate the value function, via simulation, in the context of expected infinite-horizon discounted rewards for Markov chains.