Timezone: »
In many RL applications, once training ends, it is vital to detect any deterioration in the agent performance as soon as possible. Furthermore, it often has to be done without modifying the policy and under minimal assumptions regarding the environment. In this paper, we address this problem by focusing directly on the rewards and testing for degradation. We consider an episodic framework, where the rewards within each episode are not independent, nor identically-distributed, nor Markov. We present this problem as a multivariate mean-shift detection problem with possibly partial observations. We define the mean-shift in a way corresponding to deterioration of a temporal signal (such as the rewards), and derive a test for this problem with optimal statistical power. Empirically, on deteriorated rewards in control problems (generated using various environment modifications), the test is demonstrated to be more powerful than standard tests - often by orders of magnitude. We also suggest a novel Bootstrap mechanism for False Alarm Rate control (BFAR), applicable to episodic (non-i.i.d) signal and allowing our test to run sequentially in an online manner. Our method does not rely on a learned model of the environment, is entirely external to the agent, and in fact can be applied to detect changes or drifts in any episodic signal.
Author Information
Ido Greenberg (Technion)
Shie Mannor (Technion)
Related Events (a corresponding poster, oral, or spotlight)
-
2021 Poster: Detecting Rewards Deterioration in Episodic Reinforcement Learning »
Thu. Jul 22nd 04:00 -- 06:00 PM Room Virtual
More from the Same Authors
-
2023 Poster: Representation-Driven Reinforcement Learning »
Ofir Nabati · Guy Tennenholtz · Shie Mannor -
2023 Poster: Learning to Initiate and Reason in Event-Driven Cascading Processes »
Yuval Atzmon · Eli Meirom · Shie Mannor · Gal Chechik -
2023 Poster: PPG Reloaded: An Empirical Study on What Matters in Phasic Policy Gradient »
Kaixin Wang · Zhou Daquan · Jiashi Feng · Shie Mannor -
2023 Poster: Learning Hidden Markov Models When the Locations of Missing Observations are Unknown »
BINYAMIN PERETS · Mark Kozdoba · Shie Mannor -
2023 Poster: Reward-Mixing MDPs with Few Contexts are Learnable »
Jeongyeol Kwon · Yonathan Efroni · Constantine Caramanis · Shie Mannor -
2022 Poster: Analysis of Stochastic Processes through Replay Buffers »
Shirli Di-Castro Shashua · Shie Mannor · Dotan Di Castro -
2022 Poster: Actor-Critic based Improper Reinforcement Learning »
Mohammadi Zaki · Avi Mohan · Aditya Gopalan · Shie Mannor -
2022 Poster: Optimizing Tensor Network Contraction Using Reinforcement Learning »
Eli Meirom · Haggai Maron · Shie Mannor · Gal Chechik -
2022 Poster: The Geometry of Robust Value Functions »
Kaixin Wang · Navdeep Kumar · Kuangqi Zhou · Bryan Hooi · Jiashi Feng · Shie Mannor -
2022 Spotlight: The Geometry of Robust Value Functions »
Kaixin Wang · Navdeep Kumar · Kuangqi Zhou · Bryan Hooi · Jiashi Feng · Shie Mannor -
2022 Spotlight: Actor-Critic based Improper Reinforcement Learning »
Mohammadi Zaki · Avi Mohan · Aditya Gopalan · Shie Mannor -
2022 Spotlight: Analysis of Stochastic Processes through Replay Buffers »
Shirli Di-Castro Shashua · Shie Mannor · Dotan Di Castro -
2022 Spotlight: Optimizing Tensor Network Contraction Using Reinforcement Learning »
Eli Meirom · Haggai Maron · Shie Mannor · Gal Chechik -
2022 Poster: Coordinated Attacks against Contextual Bandits: Fundamental Limits and Defense Mechanisms »
Jeongyeol Kwon · Yonathan Efroni · Constantine Caramanis · Shie Mannor -
2022 Spotlight: Coordinated Attacks against Contextual Bandits: Fundamental Limits and Defense Mechanisms »
Jeongyeol Kwon · Yonathan Efroni · Constantine Caramanis · Shie Mannor -
2021 : Invited Speaker: Shie Mannor: Lenient Regret »
Shie Mannor -
2021 : RL + Operations Research Panel »
Jim Dai · Fei Fang · Shie Mannor · Yuandong Tian · Zhiwei (Tony) Qin · Zongqing Lu -
2021 Poster: Online Limited Memory Neural-Linear Bandits with Likelihood Matching »
Ofir Nabati · Tom Zahavy · Shie Mannor -
2021 Spotlight: Online Limited Memory Neural-Linear Bandits with Likelihood Matching »
Ofir Nabati · Tom Zahavy · Shie Mannor -
2021 Poster: Confidence-Budget Matching for Sequential Budgeted Learning »
Yonathan Efroni · Nadav Merlis · Aadirupa Saha · Shie Mannor -
2021 Spotlight: Confidence-Budget Matching for Sequential Budgeted Learning »
Yonathan Efroni · Nadav Merlis · Aadirupa Saha · Shie Mannor -
2021 Poster: Value Iteration in Continuous Actions, States and Time »
Michael Lutter · Shie Mannor · Jan Peters · Dieter Fox · Animesh Garg -
2021 Spotlight: Value Iteration in Continuous Actions, States and Time »
Michael Lutter · Shie Mannor · Jan Peters · Dieter Fox · Animesh Garg -
2021 Poster: Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks »
Eli Meirom · Haggai Maron · Shie Mannor · Gal Chechik -
2021 Spotlight: Controlling Graph Dynamics with Reinforcement Learning and Graph Neural Networks »
Eli Meirom · Haggai Maron · Shie Mannor · Gal Chechik -
2020 Poster: Optimistic Policy Optimization with Bandit Feedback »
Lior Shani · Yonathan Efroni · Aviv Rosenberg · Shie Mannor -
2020 Poster: Topic Modeling via Full Dependence Mixtures »
Dan Fisher · Mark Kozdoba · Shie Mannor -
2019 Poster: Exploration Conscious Reinforcement Learning Revisited »
Lior Shani · Yonathan Efroni · Shie Mannor -
2019 Poster: Action Robust Reinforcement Learning and Applications in Continuous Control »
Chen Tessler · Chen Tessler · Yonathan Efroni · Shie Mannor -
2019 Poster: The Natural Language of Actions »
Guy Tennenholtz · Shie Mannor -
2019 Oral: Exploration Conscious Reinforcement Learning Revisited »
Lior Shani · Yonathan Efroni · Shie Mannor -
2019 Oral: The Natural Language of Actions »
Guy Tennenholtz · Shie Mannor -
2019 Poster: Nonlinear Distributional Gradient Temporal-Difference Learning »
chao qu · Shie Mannor · Huan Xu -
2019 Oral: Action Robust Reinforcement Learning and Applications in Continuous Control »
Chen Tessler · Chen Tessler · Yonathan Efroni · Yonathan Efroni · Shie Mannor · Shie Mannor -
2019 Oral: Nonlinear Distributional Gradient Temporal-Difference Learning »
chao qu · Shie Mannor · Huan Xu -
2018 Poster: Beyond the One-Step Greedy Approach in Reinforcement Learning »
Yonathan Efroni · Gal Dalal · Bruno Scherrer · Shie Mannor -
2018 Oral: Beyond the One-Step Greedy Approach in Reinforcement Learning »
Yonathan Efroni · Gal Dalal · Bruno Scherrer · Shie Mannor -
2017 Workshop: Lifelong Learning: A Reinforcement Learning Approach »
Sarath Chandar · Balaraman Ravindran · Daniel J. Mankowitz · Shie Mannor · Tom Zahavy -
2017 Poster: Consistent On-Line Off-Policy Evaluation »
Assaf Hallak · Shie Mannor -
2017 Talk: Consistent On-Line Off-Policy Evaluation »
Assaf Hallak · Shie Mannor -
2017 Poster: End-to-End Differentiable Adversarial Imitation Learning »
Nir Baram · Oron Anschel · Itai Caspi · Shie Mannor -
2017 Poster: Multi-objective Bandits: Optimizing the Generalized Gini Index »
Robert Busa-Fekete · Balazs Szorenyi · Paul Weng · Shie Mannor -
2017 Talk: End-to-End Differentiable Adversarial Imitation Learning »
Nir Baram · Oron Anschel · Itai Caspi · Shie Mannor -
2017 Talk: Multi-objective Bandits: Optimizing the Generalized Gini Index »
Robert Busa-Fekete · Balazs Szorenyi · Paul Weng · Shie Mannor