Timezone: »
We present GradientDICE for estimating the density ratio between the state distribution of the target policy and the sampling distribution in off-policy reinforcement learning. GradientDICE fixes several problems of GenDICE (Zhang et al., 2020), the current state-of-the-art for estimating such density ratios. Namely, the optimization problem in GenDICE is not a convex-concave saddle-point problem once nonlinearity in optimization variable parameterization is introduced to ensure positivity, so primal-dual algorithms are not guaranteed to find the desired solution. However, such nonlinearity is essential to ensure the consistency of GenDICE even with a tabular representation. This is a fundamental contradiction, resulting from GenDICE's original formulation of the optimization problem. In GradientDICE, we optimize a different objective from GenDICE by using the Perron-Frobenius theorem and eliminating GenDICE's use of divergence, such that nonlinearity in parameterization is not necessary for GradientDICE, which is provably convergent under linear function approximation.
Author Information
Shangtong Zhang (University of Oxford)
Bo Liu (Auburn University)
Bo Liu is a tenure-track assistant professor in the Dept. of Computer Science and Software Engineering at Auburn University. He obtained his Ph.D. from Autonomous Learning Lab at University of Massachusetts Amherst, 2015, co-led by Drs. Sridhar Mahadevan and Andrew Barto. His primary research area covers decision-making under uncertainty, human-aided machine learning, symbolic AI, trustworthiness and interpretability in machine learning, and their numerous applications to BIGDATA, autonomous driving, and healthcare informatics. In his current research, he has more than 30 publications on several notable venues, such as NIPS, UAI, AAAI, IJCAI, AAMAS, JAIR, IEEE TNNLS, ACM TECS, etc. His research is funded by NSF, Amazon, Tencent (China), Adobe, and ETRI (South Korea). He is the recipient of the UAI'2015 Facebook best student paper award and the Amazon research award in 2018. His research results have been covered by many prestigious venues, including the classical textbook "Reinforcement Learning: An Introduction" (2nd edition), NIPS'2015/IJCAI'2016/AAAI'2019 tutorials.
Shimon Whiteson (University of Oxford)
More from the Same Authors
-
2020 Poster: Provably Convergent Two-Timescale Off-Policy Actor-Critic with Function Approximation »
Shangtong Zhang · Bo Liu · Hengshuai Yao · Shimon Whiteson -
2020 Poster: Deep Coordination Graphs »
Wendelin Boehmer · Vitaly Kurin · Shimon Whiteson -
2019 Poster: Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learning »
Jakob Foerster · Francis Song · Edward Hughes · Neil Burch · Iain Dunning · Shimon Whiteson · Matthew Botvinick · Michael Bowling -
2019 Oral: Bayesian Action Decoder for Deep Multi-Agent Reinforcement Learning »
Jakob Foerster · Francis Song · Edward Hughes · Neil Burch · Iain Dunning · Shimon Whiteson · Matthew Botvinick · Michael Bowling -
2019 Poster: A Baseline for Any Order Gradient Estimation in Stochastic Computation Graphs »
Jingkai Mao · Jakob Foerster · Tim Rocktäschel · Maruan Al-Shedivat · Gregory Farquhar · Shimon Whiteson -
2019 Poster: Fast Context Adaptation via Meta-Learning »
Luisa Zintgraf · Kyriacos Shiarlis · Vitaly Kurin · Katja Hofmann · Shimon Whiteson -
2019 Oral: A Baseline for Any Order Gradient Estimation in Stochastic Computation Graphs »
Jingkai Mao · Jakob Foerster · Tim Rocktäschel · Maruan Al-Shedivat · Gregory Farquhar · Shimon Whiteson -
2019 Oral: Fast Context Adaptation via Meta-Learning »
Luisa Zintgraf · Kyriacos Shiarlis · Vitaly Kurin · Katja Hofmann · Shimon Whiteson -
2019 Poster: Fingerprint Policy Optimisation for Robust Reinforcement Learning »
Supratik Paul · Michael A Osborne · Shimon Whiteson -
2019 Oral: Fingerprint Policy Optimisation for Robust Reinforcement Learning »
Supratik Paul · Michael A Osborne · Shimon Whiteson -
2018 Poster: Fourier Policy Gradients »
Matthew Fellows · Kamil Ciosek · Shimon Whiteson -
2018 Oral: Fourier Policy Gradients »
Matthew Fellows · Kamil Ciosek · Shimon Whiteson -
2018 Poster: QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning »
Tabish Rashid · Mikayel Samvelyan · Christian Schroeder · Gregory Farquhar · Jakob Foerster · Shimon Whiteson -
2018 Poster: Deep Variational Reinforcement Learning for POMDPs »
Maximilian Igl · Luisa Zintgraf · Tuan Anh Le · Frank Wood · Shimon Whiteson -
2018 Oral: Deep Variational Reinforcement Learning for POMDPs »
Maximilian Igl · Luisa Zintgraf · Tuan Anh Le · Frank Wood · Shimon Whiteson -
2018 Oral: QMIX: Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learning »
Tabish Rashid · Mikayel Samvelyan · Christian Schroeder · Gregory Farquhar · Jakob Foerster · Shimon Whiteson -
2018 Poster: DiCE: The Infinitely Differentiable Monte Carlo Estimator »
Jakob Foerster · Gregory Farquhar · Maruan Al-Shedivat · Tim Rocktäschel · Eric Xing · Shimon Whiteson -
2018 Poster: TACO: Learning Task Decomposition via Temporal Alignment for Control »
Kyriacos Shiarlis · Markus Wulfmeier · Sasha Salter · Shimon Whiteson · Ingmar Posner -
2018 Oral: TACO: Learning Task Decomposition via Temporal Alignment for Control »
Kyriacos Shiarlis · Markus Wulfmeier · Sasha Salter · Shimon Whiteson · Ingmar Posner -
2018 Oral: DiCE: The Infinitely Differentiable Monte Carlo Estimator »
Jakob Foerster · Gregory Farquhar · Maruan Al-Shedivat · Tim Rocktäschel · Eric Xing · Shimon Whiteson -
2017 Poster: Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning »
Jakob Foerster · Nantas Nardelli · Gregory Farquhar · Triantafyllos Afouras · Phil Torr · Pushmeet Kohli · Shimon Whiteson -
2017 Talk: Stabilising Experience Replay for Deep Multi-Agent Reinforcement Learning »
Jakob Foerster · Nantas Nardelli · Gregory Farquhar · Triantafyllos Afouras · Phil Torr · Pushmeet Kohli · Shimon Whiteson