Reinforcement learning (RL) is recognized as lacking generalization and robustness under environmental perturbations, which excessively restricts its application for real-world robotics. Prior work claimed that adding regularization to the value function is equivalent to learning a robust policy with uncertain transitions. Although the regularization-robustness transformation is appealing for its simplicity and efficiency, it is still lacking in continuous control tasks. In this paper, we propose a new regularizer named $\textbf{U}$ncertainty $\textbf{S}$et $\textbf{R}$egularizer (USR), by formulating the uncertainty set on the parameter space of the transition function. In particular, USR is flexible enough to be plugged into any existing RL framework. To deal with unknown uncertainty sets, we further propose a novel adversarial approach to generate them based on the value function. We evaluate USR on the Real-world Reinforcement Learning (RWRL) benchmark, demonstrating improvements in the robust performance for perturbed testing environments.
Policy-based reinforcement learning methods suffer from the policy collapse problem. We find valued-based reinforcement learning methods with ε-greedy mechanism are capable of enjoying three characteristics, Closed-form Diversity, Objective-invariant Exploration and Adaptive Trade-off, which help value-based methods avoid the policy collapse problem. However, there does not exist a parallel mechanism for policy-based methods that achieves all three characteristics. In this paper, we propose an entropy regularization free mechanism that is designed for policy-based methods, which achieves Close…
Entropy regularization is an important idea in reinforcement learning, with great success in recent algorithms like Soft Q Network (SQN) and Soft Actor-Critic (SAC1). In this work, we extend this idea into the on-policy realm. We propose the soft policy gradient theorem (SPGT) for on-policy maximum entropy reinforcement learning. With SPGT, a series of new policy optimization algorithms are derived, such as SPG, SA2C, SA3C, SDDPG, STRPO, SPPO, SIMPALA and so on. We find that SDDPG is equivalent to SAC1. For policy gradient, the policy network is often represented as a Gaussian distribution wit…
The outcome of Jacobian singular values regularization was studied for supervised learning problems. It also was shown that Jacobian conditioning regularization can help to avoid the ``mode-collapse'' problem in Generative Adversarial Networks. In this paper, we try to answer the following question: Can information about policy conditioning help to shape a more stable and general policy of reinforcement learning agents? To answer this question, we conduct a study of Jacobian conditioning behavior during policy optimization. To the best of our knowledge, this is the first work that research con…
DeepStock: Reinforcement Learning with Policy Regularizations for Inventory Management
Booth School of Business, University of Chicago（芝加哥大学商学院）；Taobao & Tmall Group（淘宝与天猫集团）；School of Economics, Sichuan University（四川大学经济学院）；Graduate School of Business, Columbia University（哥伦比亚大学商学院）；School of Operations Research and Information Engineering, Cornell University（康奈尔大学运筹学与信息工程学院）
Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳））；University of Science and Technology of China（中国科学技术大学）；Beihang University（北京航空航天大学）
Offline reinforcement learning (RL) defines a sample-efficient learning paradigm, where a policy is learned from static and previously collected datasets without additional interaction with the environment. The major obstacle to offline RL is the estimation error arising from evaluating the value of out-of-distribution actions. To tackle this problem, most existing offline RL methods attempt to acquire a policy both ``close" to the behaviors contained in the dataset and sufficiently improved over them, which requires a trade-off between two possibly conflicting targets. In this paper, we propo…
In this paper, a sparse Markov decision process (MDP) with novel causal sparse Tsallis entropy regularization is proposed.The proposed policy regularization induces a sparse and multi-modal optimal policy distribution of a sparse MDP. The full mathematical analysis of the proposed sparse MDP is provided.We first analyze the optimality condition of a sparse MDP. Then, we propose a sparse value iteration method which solves a sparse MDP and then prove the convergence and optimality of sparse value iteration using the Banach fixed point theorem. The proposed sparse MDP is compared to soft MDPs wh…
Nara Institute of Science and Technology（奈良先端科学技术大学院大学）
In this paper, we propose cautious policy programming (CPP), a novel value-based reinforcement learning (RL) algorithm that can ensure monotonic policy improvement during learning. Based on the nature of entropy-regularized RL, we derive a new entropy regularization-aware lower bound of policy improvement that only requires estimating the expected policy advantage function. CPP leverages this lower bound as a criterion for adjusting the degree of a policy update for alleviating policy oscillation. Different from similar algorithms that are mostly theory-oriented, we also propose a novel interp…
Distributionally Robust Optimization (DRO) has enabled to prove the equivalence between robustness and regularization in classification and regression, thus providing an analytical reason why regularization generalizes well in statistical learning. Although DRO's extension to sequential decision-making overcomes $\textit{external uncertainty}$ through the robust Markov Decision Process (MDP) setting, the resulting formulation is hard to solve, especially on large domains. On the other hand, existing regularization methods in reinforcement learning only address $\textit{internal uncertainty}$ d…
Center for Artificial Intelligence Research（人工智能研究中心）；University of Agder（阿格德大学）；Sparebank 1 SR-Bank（Sparebank 1 SR银行）
The increased complexity of state-of-the-art reinforcement learning (RL) algorithms have resulted in an opacity that inhibits explainability and understanding. This has led to the development of several post-hoc explainability methods that aim to extract information from learned policies thus aiding explainability. These methods rely on empirical observations of the policy and thus aim to generalize a characterization of agents' behaviour. In this study, we have instead developed a method to imbue a characteristic behaviour into agents' policies through regularization of their objective functi…
Learning in sparse reward settings remains a challenge in Reinforcement Learning, which is often addressed by using intrinsic rewards. One promising strategy is inspired by human curiosity, requiring the agent to learn to predict the future. In this paper a curiosity-driven agent is extended to use these predictions directly for training. To achieve this, the agent predicts the value function of the next state at any point in time. Subsequently, the consistency of this prediction with the current value function is measured, which is then used as a regularization term in the loss function of th…
Nara Institute of Science and Technology（奈良先端科学技术大学院大学）；Osaka University（大阪大学）；Advanced Telecommunication Research（国际电气通信基础技术研究所）
Maximum Tsallis entropy (MTE) framework in reinforcement learning has gained popularity recently by virtue of its flexible modeling choices including the widely used Shannon entropy and sparse entropy. However, non-Shannon entropies suffer from approximation error and subsequent underperformance either due to its sensitivity or the lack of closed-form policy expression. To improve the tradeoff between flexibility and empirical performance, we propose to strengthen their error-robustness by enforcing implicit Kullback-Leibler (KL) regularization in MTE motivated by Munchausen DQN (MDQN). We do …
Department of Applied Mathematics, Harvard University（哈佛大学应用数学系）；Department of Computer Science, Harvard University（哈佛大学计算机科学系）；Department of Statistics, Harvard University（哈佛大学统计学系）
In batch reinforcement learning, there can be poorly explored state-action pairs resulting in poorly learned, inaccurate models and poorly performing associated policies. Various regularization methods can mitigate the problem of learning overly-complex models in Markov decision processes (MDPs), however they operate in technically and intuitively distinct ways and lack a common form in which to compare them. This paper unifies three regularization methods in a common framework -- a weighted average transition matrix. Considering regularization methods in this common form illuminates how the M…
New York University（纽约大学）；Google Research（谷歌研究院）；Google DeepMind（谷歌DeepMind）
Many modern approaches to offline Reinforcement Learning (RL) utilize behavior regularization, typically augmenting a model-free actor critic algorithm with a penalty measuring divergence of the policy from the offline data. In this work, we propose an alternative approach to encouraging the learned policy to stay close to the data, namely parameterizing the critic as the log-behavior-policy, which generated the offline data, plus a state-action value offset term, which can be learned using a neural network. Behavior regularization then corresponds to an appropriate regularizer on the offset t…
CNRS (French National Center for Scientific Research)（法国国家科学研究中心）；Univ. Grenoble Alpes（格勒诺布尔阿尔卑斯大学）；University of Wisconsin（威斯康星大学）
We investigate a class of reinforcement learning dynamics where players adjust their strategies based on their actions' cumulative payoffs over time - specifically, by playing mixed strategies that maximize their expected cumulative payoff minus a regularization term. A widely studied example is exponential reinforcement learning, a process induced by an entropic regularization term which leads mixed strategies to evolve according to the replicator dynamics. However, in contrast to the class of regularization functions used to define smooth best responses in models of stochastic fictitious pla…
This paper proposes a new optimization objective for value-based deep reinforcement learning. We extend conventional Deep Q-Networks (DQNs) by adding a model-learning component yielding a transcoder network. The prediction errors for the model are included in the basic DQN loss as additional regularizers. This augmented objective leads to a richer training signal that provides feedback at every time step. Moreover, because learning an environment model shares a common structure with the RL problem, we hypothesize that the resulting objective improves both sample efficiency and performance. We …
University of Waterloo（滑铁卢大学）；Columbia University（哥伦比亚大学）
We propose \emph{Choquet regularizers} to measure and manage the level of exploration for reinforcement learning (RL), and reformulate the continuous-time entropy-regularized RL problem of Wang et al. (2020, JMLR, 21(198)) in which we replace the differential entropy used for regularization with a Choquet regularizer. We derive the Hamilton--Jacobi--Bellman equation of the problem, and solve it explicitly in the linear--quadratic (LQ) case via maximizing statically a mean--variance constrained Choquet regularizer. Under the LQ setting, we derive explicit optimal distributions for several speci…
JD Intelligent Cities Research, JD Technology（京东科技京东智能城市研究院）；Institute for AI Industry Research (AIR), Tsinghua University（清华大学人工智能产业研究院）
Most prior approaches to offline reinforcement learning (RL) utilize \textit{behavior regularization}, typically augmenting existing off-policy actor critic algorithms with a penalty measuring divergence between the policy and the offline data. However, these approaches lack guaranteed performance improvement over the behavior policy. In this work, we start from the performance difference between the learned policy and the behavior policy, we derive a new policy learning objective that can be used in the offline setting, which corresponds to the advantage function value of the behavior policy,…
University of Alberta（阿尔伯塔大学）；Google Research（谷歌研究院）；DeepMind Alberta（DeepMind 阿尔伯塔）
Deep reinforcement learning algorithms have shown an impressive ability to learn complex control policies in high-dimensional tasks. However, despite the ever-increasing performance on popular benchmarks, policies learned by deep reinforcement learning algorithms can struggle to generalize when evaluated in remarkably similar environments. In this paper we propose a protocol to evaluate generalization in reinforcement learning through different modes of Atari 2600 games. With that protocol we assess the generalization capabilities of DQN, one of the most traditional deep reinforcement learning…
University of California at Berkeley（加州大学伯克利分校）；Toyota Technological Institute at Chicago（芝加哥丰田理工学院）
We provide an algorithm that achieves the optimal regret rate in an unknown weakly communicating Markov Decision Process (MDP). The algorithm proceeds in episodes where, in each episode, it picks a policy using regularization based on the span of the optimal bias vector. For an MDP with S states and A actions whose optimal bias vector has span bounded by H, we show a regret bound of ~O(HSpAT). We also relate the span to various diameter-like quantities associated with the MDP, demonstrating how our results improve on previous regret bounds.
We consider the problem of detecting out-of-distribution (OOD) samples in deep reinforcement learning. In a value based reinforcement learning setting, we propose to use uncertainty estimation techniques directly on the agent's value estimating neural network to detect OOD samples. The focus of our work lies in analyzing the suitability of approximate Bayesian inference methods and related ensembling techniques that generate uncertainty estimates. Although prior work has shown that dropout-based variational inference techniques and bootstrap-based approaches can be used to model epistemic unce…
Technion – Israel Institute of Technology（以色列理工学院）
In the Bayesian reinforcement learning (RL) setting, a prior distribution over the unknown problem parameters -- the rewards and transitions -- is assumed, and a policy that optimizes the (posterior) expected return is sought. A common approximation, which has been recently popularized as meta-RL, is to train the agent on a sample of $N$ problem instances from the prior, with the hope that for large enough $N$, good generalization behavior to an unseen test instance will be obtained. In this work, we study generalization in Bayesian RL under the probably approximately correct (PAC) framework, …
All Models are Wrong, Knowing Where is Useful: On Model Uncertainty in Reinforcement Learning
German Federal Ministry of Research, Technology and Space (BMFTR)（德国联邦研究、技术和空间部）；Robotics Institute Germany (RIG)（德国机器人研究所）；Institute for Data Science in Mechanical Engineering, RWTH Aachen University（机械工程数据科学研究所，亚琛工业大学）；NHR Center NHR4CES at RWTH Aachen University（亚琛工业大学NHR4CES中心）
Uncertainty is ubiquitous in games, both in the agents playing games and often in the games themselves. Working with uncertainty is therefore an important component of successful deep reinforcement learning agents. While there has been substantial effort and progress in understanding and working with uncertainty for supervised learning, the body of literature for uncertainty aware deep reinforcement learning is less developed. While many of the same problems regarding uncertainty in neural networks for supervised learning remain for reinforcement learning, there are additional sources of uncer…
In this paper, the issue of model uncertainty in safety-critical control is addressed with a data-driven approach. For this purpose, we utilize the structure of an input-ouput linearization controller based on a nominal model along with a Control Barrier Function and Control Lyapunov Function based Quadratic Program (CBF-CLF-QP). Specifically, we propose a novel reinforcement learning framework which learns the model uncertainty present in the CBF and CLF constraints, as well as other control-affine dynamic constraints in the quadratic program. The trained policy is combined with the nominal m…
We propose a simple architecture for deep reinforcement learning by embedding inputs into a learned Fourier basis and show that it improves the sample efficiency of both state-based and image-based RL. We perform infinite-width analysis of our architecture using the Neural Tangent Kernel and theoretically show that tuning the initial variance of the Fourier basis is equivalent to functional regularization of the learned deep network. That is, these learned Fourier features allow for adjusting the degree to which networks underfit or overfit different frequencies in the training data, and hence…
Technion Institute of Technology（以色列理工学院）；Nvidia Research（英伟达研究院）
Model-based offline reinforcement learning approaches generally rely on bounds of model error. Estimating these bounds is usually achieved through uncertainty estimation methods. In this work, we combine parametric and nonparametric methods for uncertainty estimation through a novel latent space based metric. In particular, we build upon recent advances in Riemannian geometry of generative models to construct a pullback metric of an encoder-decoder based forward model. Our proposed metric measures both the quality of out-of-distribution samples as well as the discrepancy of examples in the dat…
Reinforcement learning agents are faced with two types of uncertainty. Epistemic uncertainty stems from limited data and is useful for exploration, whereas aleatoric uncertainty arises from stochastic environments and must be accounted for in risk-sensitive applications. We highlight the challenges involved in simultaneously estimating both of them, and propose a framework for disentangling and estimating these uncertainties on learned Q-values. We derive unbiased estimators of these uncertainties and introduce an uncertainty-aware DQN algorithm, which we show exhibits safe learning behavior a…
German Research Center for AI (DFKI)（德国人工智能研究中心（DFKI））；Centre for Cognitive Science（认知科学中心）
Recent methods for imitation learning directly learn a $Q$-function using an implicit reward formulation rather than an explicit reward function. However, these methods generally require implicit reward regularization to improve stability and often mistreat absorbing states. Previous works show that a squared norm regularization on the implicit reward function is effective, but do not provide a theoretical analysis of the resulting properties of the algorithms. In this work, we show that using this regularizer under a mixture distribution of the policy and the expert provides a particularly il…
Nara Institute of Science and Technology（奈良先端科学技术大学院大学）
The recent boom in the literature on entropy-regularized reinforcement learning (RL) approaches reveals that Kullback-Leibler (KL) regularization brings advantages to RL algorithms by canceling out errors under mild assumptions. However, existing analyses focus on fixed regularization with a constant weighting coefficient and do not consider cases where the coefficient is allowed to change dynamically. In this paper, we study the dynamic coefficient scheme and present the first asymptotic error bound. Based on the dynamic coefficient error bound, we propose an effective scheme to tune the coef…
Offline Multi-Agent Reinforcement Learning with Implicit Global-to-Local Value Regularization
Institute for AI Industry Research (AIR), Tsinghua University（清华大学智能产业研究院（AIR））；Beijing Jiaotong University（北京交通大学）；UT Austin（德克萨斯大学奥斯汀分校）；Tsinghua University（清华大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）
Viability of Future Actions: Robust Safety in Reinforcement Learning via Entropy Regularization
Institute for Data Science in Mechanical Engineering, RWTH Aachen University, Germany（机械工程数据科学研究所，亚琛RWTH大学，德国）；Learning Systems and Robotics Lab, Technical University of Munich, Germany（学习系统与机器人实验室，慕尼黑技术大学，德国）
