Timezone: »
In this paper we address the problem of decision making within a Markov decision process (MDP) framework where risk and modeling errors are taken into account. Our approach is to minimize a risk-sensitive conditional-value-at-risk (CVaR) objective, as opposed to a standard risk-neutral expectation. We refer to such problem as CVaR MDP. Our first contribution is to show that a CVaR objective, besides capturing risk sensitivity, has an alternative interpretation as expected cost under worst-case modeling errors, for a given error budget. This result, which is of independent interest, motivates CVaR MDPs as a unifying framework for risk-sensitive and robust decision making. Our second contribution is to present a value-iteration algorithm for CVaR MDPs, and analyze its convergence rate. To our knowledge, this is the first solution algorithm for CVaR MDPs that enjoys error guarantees. Finally, we present results from numerical experiments that corroborate our theoretical findings and show the practicality of our approach.
Author Information
Yinlam Chow (Stanford)
Aviv Tamar (UC Berkeley)
Shie Mannor (Technion)
Marco Pavone (Stanford University)
More from the Same Authors
-
2020 Poster: Continuous Meta-Learning without Tasks »
James Harrison · Apoorva Sharma · Chelsea Finn · Marco Pavone -
2020 Poster: Evidential Sparsification of Multimodal Latent Spaces in Conditional Variational Autoencoders »
Masha Itkina · Boris Ivanovic · Ransalu Senanayake · Mykel J Kochenderfer · Marco Pavone -
2019 Workshop: Safety and Robustness in Decision-making »
Mohammad Ghavamzadeh · Shie Mannor · Yisong Yue · Marek Petrik · Yinlam Chow -
2019 Poster: High-Dimensional Optimization in Adaptive Random Subspaces »
Jonathan Lacotte · Mert Pilanci · Marco Pavone -
2019 Poster: Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies »
Yonathan Efroni · Nadav Merlis · Mohammad Ghavamzadeh · Shie Mannor -
2019 Spotlight: Tight Regret Bounds for Model-Based Reinforcement Learning with Greedy Policies »
Yonathan Efroni · Nadav Merlis · Mohammad Ghavamzadeh · Shie Mannor -
2018 Poster: Multiple-Step Greedy Policies in Approximate and Online Reinforcement Learning »
Yonathan Efroni · Gal Dalal · Bruno Scherrer · Shie Mannor -
2018 Spotlight: Multiple-Step Greedy Policies in Approximate and Online Reinforcement Learning »
Yonathan Efroni · Gal Dalal · Bruno Scherrer · Shie Mannor -
2017 Poster: Rotting Bandits »
Nir Levine · Yacov Crammer · Shie Mannor -
2017 Poster: Shallow Updates for Deep Reinforcement Learning »
Nir Levine · Tom Zahavy · Daniel J Mankowitz · Aviv Tamar · Shie Mannor -
2016 Poster: Safe Policy Improvement by Minimizing Robust Baseline Regret »
Mohammad Ghavamzadeh · Marek Petrik · Yinlam Chow -
2016 Poster: Adaptive Skills Adaptive Partitions (ASAP) »
Daniel J Mankowitz · Timothy A Mann · Shie Mannor -
2015 Workshop: Machine Learning for (e-)Commerce »
Esteban Arcaute · Mohammad Ghavamzadeh · Shie Mannor · Georgios Theocharous -
2015 Poster: Online Learning for Adversaries with Memory: Price of Past Mistakes »
Oren Anava · Elad Hazan · Shie Mannor -
2015 Poster: Policy Gradient for Coherent Risk Measures »
Aviv Tamar · Yinlam Chow · Mohammad Ghavamzadeh · Shie Mannor -
2015 Poster: Community Detection via Measure Space Embedding »
Mark Kozdoba · Shie Mannor -
2014 Workshop: From Bad Models to Good Policies (Sequential Decision Making under Uncertainty) »
Odalric-Ambrym Maillard · Timothy A Mann · Shie Mannor · Jeremie Mary · Laurent Orseau · Thomas Dietterich · Ronald Ortner · Peter Grünwald · Joelle Pineau · Raphael Fonteneau · Georgios Theocharous · Esteban D Arcaute · Christos Dimitrakakis · Nan Jiang · Doina Precup · Pierre-Luc Bacon · Marek Petrik · Aviv Tamar -
2014 Poster: "How hard is my MDP?" The distribution-norm to the rescue »
Odalric-Ambrym Maillard · Timothy A Mann · Shie Mannor -
2014 Poster: Robust Logistic Regression and Classification »
Jiashi Feng · Huan Xu · Shie Mannor · Shuicheng Yan -
2014 Oral: "How hard is my MDP?" The distribution-norm to the rescue »
Odalric-Ambrym Maillard · Timothy A Mann · Shie Mannor -
2014 Poster: Algorithms for CVaR Optimization in MDPs »
Yinlam Chow · Mohammad Ghavamzadeh -
2013 Poster: Reinforcement Learning in Robust Markov Decision Processes »
Shiau Hong Lim · Huan Xu · Shie Mannor -
2013 Poster: Online PCA for Contaminated Data »
Jiashi Feng · Huan Xu · Shie Mannor · Shuicheng Yan -
2013 Poster: Learning Multiple Models via Regularized Weighting »
Daniel Vainsencher · Shie Mannor · Huan Xu -
2012 Poster: The Perturbed Variation »
Maayan Harel · Shie Mannor -
2011 Poster: From Bandits to Experts: On the Value of Side-Observations »
Shie Mannor · Ohad Shamir -
2011 Spotlight: From Bandits to Experts: On the Value of Side-Observations »
Shie Mannor · Ohad Shamir -
2011 Poster: Committing Bandits »
Loc X Bui · Ramesh Johari · Shie Mannor -
2010 Spotlight: Online Classification with Specificity Constraints »
Andrey Bernstein · Shie Mannor · Nahum Shimkin -
2010 Poster: Online Classification with Specificity Constraints »
Andrey Bernstein · Shie Mannor · Nahum Shimkin -
2010 Poster: Distributionally Robust Markov Decision Processes »
Huan Xu · Shie Mannor