paper-with-me

홈 › Papers

The Online Coupon-Collector Problem and Its Application to Lifelong Reinforcement Learning

2015-06-10 · Emma Brunskill, Lihong Li

Transferring knowledge across a sequence of related tasks is an important challenge in reinforcement learning (RL). Despite much encouraging empirical evidence, there has been little theoretical analysis. In this paper, we study a class of lifelong RL problems: the agent solves a sequence of tasks modeled as finite Markov decision processes (MDPs), each of which is from a finite set of MDPs with the same state/action sets and different transition/reward functions. Motivated by the need for cross-task exploration in lifelong learning, we formulate a novel online coupon-collector problem and give an optimal algorithm. This allows us to develop a new lifelong RL algorithm, whose overall sample complexity in a sequence of tasks is much smaller than single-task learning, even if the sequence of tasks is generated by an adversary. Benefits of the algorithm are demonstrated in simulated problems, including a recently introduced human-robot interaction problem.

📄 PDF Abstract BibTeX arXiv:1506.03379

Code (0)

등록된 구현이 없습니다.

Tasks

Lifelong learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Experimental quantum advantage with quantum coupon collector

2021-12-15 · Min-Gang Zhou, Xiao-Yu Cao, Yu-Shuo Lu, Yang Wang 외

An increasing number of communication and computational schemes with quantum advantages have recently been proposed, which implies that quantum technology has fertile application prospects. However, demonstrating these s…

Collecting Coupons with Random Initial Stake

2013-08-29 · Benjamin Doerr, Carola Doerr

Motivated by a problem in the theory of randomized search heuristics, we give a very precise analysis for the coupon collector problem where the collector starts with a random set of coupons (chosen uniformly from all se…

Proper vs Improper Quantum PAC learning

2024-03-05 · Ashwin Nayak, Pulkit Sinha

A basic question in the PAC model of learning is whether proper learning is harder than improper learning. In the classical case, there are examples of concept classes with VC dimension $d$ that have sample complexity $\…

PAC learning

Column Bound for Orthogonal Matrix Factorization

2024-05-21 · Anirudh Dash

This article explores the intersection of the Coupon Collector's Problem and the Orthogonal Matrix Factorization (OMF) problem. Specifically, we derive bounds on the minimum number of columns $p$ (in $\mathbf{X}$) requir…

Optimal lower bounds for Quantum Learning via Information Theory

2023-01-05 · Shima Bab Hadiashar, Ashwin Nayak, Pulkit Sinha

Although a concept class may be learnt more efficiently using quantum samples as compared with classical samples in certain scenarios, Arunachalam and de Wolf (JMLR, 2018) proved that quantum learners are asymptotically …

Learning TheoryPAC learning