Time-Efficient Reinforcement Learning with Stochastic Stateful Policies
Firas Al-Hafez, Guoping Zhao, Jan Peters, Davide Tateo
Abstract
Stateful policies play an important role in reinforcement learning, such as handling partially observable environments, enhancing robustness, or imposing an inductive bias directly into the policy structure. The conventional method for training stateful policies is Backpropagation Through Time (BPTT), which comes with significant drawbacks, such as slow training due to sequential gradient propagation and the occurrence of vanishing or exploding gradients. The gradient is often truncated to address these issues, resulting in a biased policy update. We present a novel approach for training stateful policies by decomposing the latter into a stochastic internal state kernel and a stateless policy, jointly optimized by following the stateful policy gradient. We introduce different versions of the stateful policy gradient theorem, enabling us to easily instantiate stateful variants of popular reinforcement learning and imitation learning algorithms. Furthermore, we provide a theoretical analysis of our new gradient estimator and compare it with BPTT. We evaluate our approach on complex continuous control tasks, e.g., humanoid locomotion, and demonstrate that our gradient estimator scales effectively with task complexity while offering a faster and simpler alternative to BPTT.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
1 top-tier paper has cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Real-Time Recurrent Learning using Trace Units in Reinforcement Learning cites Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPs
- Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPs cites Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable Model
Cited by top-tier papers1
Builds on5
- Stochastic Latent Actor-Critic: Deep Reinforcement Learning with a Latent Variable ModelAlex X. Lee, Anusha Nagabandi, Pieter Abbeel, Sergey LevineNeurIPS 2020 · 437 citations
- Supervised Pretraining Can Learn In-Context Reinforcement LearningJonathan Lee, Annie Xie, Aldo Pacchiano, Yash Chandak et al.NeurIPS 2023 · 170 citations
- Recurrent Model-Free RL Can Be a Strong Baseline for Many POMDPsTianwei Ni, Benjamin Eysenbach, Ruslan SalakhutdinovICML 2022 · 162 citations
- Provable Reinforcement Learning with a Short-Term MemoryYonathan Efroni, Chi Jin, Akshay Krishnamurthy, Sobhan MiryoosefiICML 2022 · 45 citations
- LS-IQ: Implicit Reward Regularization for Inverse Reinforcement LearningFiras Al-Hafez, Davide Tateo, Oleg Arenz, Guoping Zhao et al.ICLR 2023 · 1 citation
Related papers
- Training Recurrent Neural Networks Online by Learning Explicit State VariablesSomjit Nath, Vincent Liu, Alan Chan, Xin Li et al.ICLR 2020 · 9 citations
- Learning Value Functions in Deep Policy Gradients using Residual VarianceYannis Flet-Berliac, Reda Ouhamma, Odalric-Ambrym Maillard, Philippe PreuxICLR 2021 · 16 citations
- Synthesizing Programmatic Policies that Inductively GeneralizeJeevana Priya Inala, Osbert Bastani, Zenna Tavares, Armando Solar-LezamaICLR 2020 · 54 citations
- Stabilizing Backpropagation Through Time to Learn Complex PhysicsPatrick Schnell, Nils ThuereyICLR 2024 · 2 citations
- Reparameterization Flow Policy OptimizationHai Zhong, Zhuoran Li, Xun Wang, Longbo HuangICML 2026
