Fast EXP3 Algorithms
Machine Learning
2025-12-15 v1 Artificial Intelligence
Data Structures and Algorithms
Abstract
We point out that EXP3 can be implemented in constant time per round, propose more practical algorithms, and analyze the trade-offs between the regret bounds and time complexities of these algorithms.
Cite
@article{arxiv.2512.11201,
title = {Fast EXP3 Algorithms},
author = {Ryoma Sato and Shinji Ito},
journal= {arXiv preprint arXiv:2512.11201},
year = {2025}
}
Related papers
View all related →
Machine Learning · Computer Science
Efficient and Robust Algorithms for Adversarial Linear Contextual Bandits
Gergely Neu, Julia Olkhovskaya
2022-05-25
Machine Learning · Computer Science
Adapting to Delays and Data in Adversarial Multi-Armed Bandits
András György, Pooria Joulani
2020-10-14
Machine Learning · Computer Science
Improved Regret Bounds for Tracking Experts with Memory
James Robinson, Mark Herbster
2021-06-25
Optimization and Control · Mathematics
Anytime Proximity Moving Horizon Estimation: Stability and Regret
Meriem Gharbi, Bahman Gharesifard, Christian Ebenbauer
2021-11-09
Data Structures and Algorithms · Computer Science
Prediction strategies without loss
Michael Kapralov, Rina Panigrahy
2012-10-11
Machine Learning · Computer Science
Regret Bounds and Reinforcement Learning Exploration of EXP-based Algorithms
Mengfan Xu, Diego Klabjan
2024-05-07
Machine Learning · Computer Science
Improved Corruption Robust Algorithms for Episodic Reinforcement Learning
Yifang Chen, Simon S. Du, Kevin Jamieson
2021-03-09
Machine Learning · Computer Science
A Tractable Algorithm For Finite-Horizon Continuous Reinforcement Learning
Phanideep Gampa, Sairam Satwik Kondamudi, Lakshmanan Kailasam
2019-08-05
Machine Learning · Computer Science
The Interplay Between Stability and Regret in Online Learning
Ankan Saha, Prateek Jain, Ambuj Tewari
2012-11-28
Machine Learning · Computer Science
Temporal Variability in Implicit Online Learning
Nicolò Campolongo, Francesco Orabona
2020-11-10
Computer Science and Game Theory · Computer Science
Balancing Adaptability and Non-exploitability in Repeated Games
Anthony DiGiovanni, Ambuj Tewari
2022-07-05
Machine Learning · Computer Science
An Improved Parametrization and Analysis of the EXP3++ Algorithm for Stochastic and Adversarial Bandits
Yevgeny Seldin, Gábor Lugosi
2017-05-10
Machine Learning · Computer Science
Combinatorial Bandits Revisited
Richard Combes, M. Sadegh Talebi, Alexandre Proutiere, Marc Lelarge
2015-11-09
Machine Learning · Statistics
Efficient tracking of a growing number of experts
Jaouad Mourtada, Odalric-Ambrym Maillard
2017-09-01
Machine Learning · Computer Science
Optimal amortized regret in every interval
Rina Panigrahy, Preyas Popat
2013-04-30
Statistics Theory · Mathematics
Improved Second-Order Bounds for Prediction with Expert Advice
Nicolo Cesa-Bianchi, Yishay Mansour, Gilles Stoltz
2007-06-13
Machine Learning · Statistics
Adapting to Non-stationarity with Growing Expert Ensembles
Cosma Rohilla Shalizi, Abigail Z. Jacobs, Kristina Lisa Klinkner, Aaron Clauset
2011-06-30
Machine Learning · Computer Science
Nonstochastic Multiarmed Bandits with Unrestricted Delays
Tobias Sommer Thune, Nicolò Cesa-Bianchi, Yevgeny Seldin
2019-11-20
Computer Science and Game Theory · Computer Science
Fast swap regret minimization and applications to approximate correlated equilibria
Binghui Peng, Aviad Rubinstein
2023-11-15
Computer Science and Game Theory · Computer Science
Uncoupled Bandit Learning towards Rationalizability: Benchmarks, Barriers, and Algorithms
Jibang Wu, Haifeng Xu, Fan Yao
2023-12-27
Machine Learning · Computer Science
A Meta-Learning Control Algorithm with Provable Finite-Time Guarantees
Deepan Muthirayan, Pramod Khargonekar
2022-02-07
Machine Learning · Computer Science
Improved Sleeping Bandits with Stochastic Actions Sets and Adversarial Rewards
Aadirupa Saha, Pierre Gaillard, Michal Valko
2020-08-11
Machine Learning · Computer Science
No Weighted-Regret Learning in Adversarial Bandits with Delays
Ilai Bistritz, Zhengyuan Zhou, Xi Chen, Nicholas Bambos +1
2022-05-16
Machine Learning · Computer Science
Distributed No-Regret Learning for Multi-Stage Systems with End-to-End Bandit Feedback
I-Hong Hou
2024-08-20
Computational Complexity · Computer Science
General Algorithms for Testing the Ambiguity of Finite Automata
Cyril Allauzen, Mehryar Mohri, Ashish Rastogi
2008-02-25