A unified framework for stochastic optimization

WB Powell - European Journal of Operational Research, 2019 - Elsevier
Stochastic optimization is an umbrella term that includes over a dozen fragmented
communities, using a patchwork of sometimes overlapping notational systems with …

The statistical complexity of interactive decision making

DJ Foster, SM Kakade, J Qian, A Rakhlin - arXiv preprint arXiv:2112.13487, 2021 - arxiv.org
A fundamental challenge in interactive learning and decision making, ranging from bandit
problems to reinforcement learning, is to provide sample-efficient, adaptive learning …

Machine learning testing: Survey, landscapes and horizons

JM Zhang, M Harman, L Ma… - IEEE Transactions on …, 2020 - ieeexplore.ieee.org
This paper provides a comprehensive survey of techniques for testing machine learning
systems; Machine Learning Testing (ML testing) research. It covers 144 papers on testing …

Introduction to multi-armed bandits

A Slivkins - Foundations and Trends® in Machine Learning, 2019 - nowpublishers.com
Multi-armed bandits a simple but very powerful framework for algorithms that make
decisions over time under uncertainty. An enormous body of work has accumulated over the …

Hyperband: A novel bandit-based approach to hyperparameter optimization

L Li, K Jamieson, G DeSalvo, A Rostamizadeh… - Journal of Machine …, 2018 - jmlr.org
Performance of machine learning algorithms depends critically on identifying a good set of
hyperparameters. While recent approaches use Bayesian optimization to adaptively select …

Reward-free exploration for reinforcement learning

C Jin, A Krishnamurthy… - … on Machine Learning, 2020 - proceedings.mlr.press
Exploration is widely regarded as one of the most challenging aspects of reinforcement
learning (RL), with many naive approaches succumbing to exponential sample complexity …

Leveraging offline data in online reinforcement learning

A Wagenmaker, A Pacchiano - International Conference on …, 2023 - proceedings.mlr.press
Two central paradigms have emerged in the reinforcement learning (RL) community: online
RL and offline RL. In the online RL setting, the agent has no prior knowledge of the …

Non-stochastic best arm identification and hyperparameter optimization

K Jamieson, A Talwalkar - Artificial intelligence and statistics, 2016 - proceedings.mlr.press
Motivated by the task of hyperparameter optimization, we introduce the\em non-stochastic
best-arm identification problem. We identify an attractive algorithm for this setting that makes …

Time-uniform, nonparametric, nonasymptotic confidence sequences

SR Howard, A Ramdas, J McAuliffe, J Sekhon - 2021 - projecteuclid.org
Time-uniform, nonparametric, nonasymptotic confidence sequences Page 1 The Annals of
Statistics 2021, Vol. 49, No. 2, 1055–1080 https://doi.org/10.1214/20-AOS1991 © Institute of …

Optimal best arm identification with fixed confidence

A Garivier, E Kaufmann - Conference on Learning Theory, 2016 - proceedings.mlr.press
We give a complete characterization of the complexity of best-arm identification in one-
parameter bandit problems. We prove a new, tight lower bound on the sample complexity …