Stochastic Decision Horizons for Survival-Based Constrained Reinforcement Learning
Max Planck Institute for Human Cognitive and Brain Sciences（马克斯·普朗克人类认知与脑科学研究所）；Center for Scalable Data Analytics and Artificial Intelligence (ScaDS.AI)（可扩展数据分析与人工智能中心 (ScaDS.AI)）；Hertie Institute for Clinical Brain Research & Center for Integrative Neuroscience（赫尔特临床脑研究所在线及整合神经科学中心）；University of Tübingen（图宾根大学）；Max Planck Institute for Intelligent Systems（马克斯·普朗克智能系统研究所）
Nikola Milosevic, Leonard Franz, Daniel Haeufle, Georg Martius, Nico Scherf, Pavel Kolev
We propose stochastic decision horizons (SDH), a theoretically grounded framework for survival-based constrained reinforcement learning from per-step violation signals. Rather than treating violations as expenditure of a cumulative budget, SDH shortens the effective decision horizon when violations occur, reducing current reward credit and future bootstrapping. We show that this construction represents a geometric-horizon survival constraint and, for a fixed Lagrange multiplier, remains the same SDH problem up to a constant reward offset that acts as an implicit incentive for survival. Using control as inference, we derive two off-policy and replay-compatible, regularized algorithms with post-violation semantics. Absorbing-state ends the decision process, so only surviving decisions pay policy KL, yielding AS-SAC. Virtual termination stops reward credit while the process continues, retaining ordinary discounted KL and yielding VT-MPO. To connect SDH with cumulative-cost CMDPs, we introduce violation-depth profiles. Under exponential-tail single-scale profiles, the SDH statistic determines cumulative cost; departures from this form mark where the connection breaks. Experiments validate both the method and this predicted scope. On the 90-muscle H2190 humanoid in Hyfydy, VT-MPO matches the EWA reference in peak gait realism, with its highest observed gait-match checkpoint at 21M environment steps versus 90M for EWA and markedly more stable training. On Safety Gymnasium, violation-depth profiles predict the regimes in which SDH achieves strong reward-violation trade-offs.
Vilcek Institute of Graduate Biomedical Sciences, NYU Grossman School of Medicine（纽约大学格罗斯曼医学院维尔切克生物医学科学研究生院）；NYU Langone Health System（纽约大学朗格尼医疗系统）；Merck & Co., Inc（默克公司）；New York University（纽约大学）
The prognosis for patients with epithelial ovarian cancer remains dismal despite improvements in survival for other cancers. Treatment involves multiple lines of chemotherapy and becomes increasingly heterogeneous after first-line therapy. Reinforcement learning with real-world outcomes data has the potential to identify novel treatment strategies to improve overall survival. We design a reinforcement learning environment to model epithelial ovarian cancer treatment trajectories and use model free reinforcement learning to investigate therapeutic regimens for simulated patients.
There has been considerable interest in leveraging RL and stochastic control methods to learn optimal treatment strategies for critically ill patients, directly from observational data. However, there is significant ambiguity on the control objective and on the best reward choice for the standard RL objective. In this work, we propose a clinically motivated control objective for critically ill patients, for which the value functions have a simple medical interpretation. Further, we present theoretical results and adapt our method to a practical Deep RL algorithm, which can be used alongside an…
Survival analysis is playing a major role in manufacturing sector by analyzing occurrence of any unwanted event based on the input data. Predictive maintenance, which is a part of survival analysis, helps to find any device failure based on the current incoming data from different sensor or any equipment. Deep learning techniques were used to automate the predictive maintenance problem to some extent, but they are not very helpful in predicting the device failure for the input data which the algorithm had not learned. Since neural network predicts the output based on previous learned input fea…
CleanSurvival: Automated data preprocessing for time-to-event models using reinforcement learning
German University in Cairo（埃及德国亚历山大大学）；German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心（DFKI））；University of Saarland（萨尔大学）；University of Kaiserslautern–Landau (RPTU)（凯撒斯劳滕-兰道大学（RPTU））
Although recent model-free reinforcement learning algorithms have been shown to be capable of mastering complicated decision-making tasks, the sample complexity of these methods has remained a hurdle to utilizing them in many real-world applications. In this regard, model-based reinforcement learning proposes some remedies. Yet, inherently, model-based methods are more computationally expensive and susceptible to sub-optimality. One reason is that model-generated data are always less accurate than real data, and this often leads to inaccurate transition and reward function models. With the aim…
Tsinghua University（清华大学）；University of Washington（华盛顿大学）
This paper gives the first polynomial-time algorithm for tabular Markov Decision Processes (MDP) that enjoys a regret bound \emph{independent on the planning horizon}. Specifically, we consider tabular MDP with $S$ states, $A$ actions, a planning horizon $H$, total reward bounded by $1$, and the agent plays for $K$ episodes. We design an algorithm that achieves an $O\left(\mathrm{poly}(S,A,\log K)\sqrt{K}\right)$ regret in contrast to existing bounds which either has an additional $\mathrm{polylog}(H)$ dependency~\citep{zhang2020reinforcement} or has an exponential dependency on $S$~\citep{li2…
We study session-based recommendation scenarios where we want to recommend items to users during sequential interactions to improve their long-term utility. Optimizing a long-term metric is challenging because the learning signal (whether the recommendations achieved their desired goals) is delayed and confounded by other user interactions with the system. Targeting immediately measurable proxies such as clicks can lead to suboptimal recommendations due to misalignment with the long-term metric. We develop a new reinforcement learning algorithm called Short Horizon Policy Improvement (SHPI) th…
University of Alberta（阿尔伯塔大学）；University of Toronto（多伦多大学）；Huawei Technologies Canada, Ltd.（加拿大华为技术有限公司）
We explore fixed-horizon temporal difference (TD) methods, reinforcement learning algorithms for a new kind of value function that predicts the sum of rewards over a $\textit{fixed}$ number of future time steps. To learn the value function for horizon $h$, these algorithms bootstrap from the value function for horizon $h-1$, or some shorter horizon. Because no value function bootstraps from itself, fixed-horizon methods are immune to the stability problems that plague other off-policy TD methods using function approximation (also known as "the deadly triad"). Although fixed-horizon methods req…
Carnegie Mellon University（卡内基梅隆大学）；Georgia Institute of Technology（佐治亚理工学院）
In this paper, we propose to combine imitation and reinforcement learning via the idea of reward shaping using an oracle. We study the effectiveness of the near-optimal cost-to-go oracle on the planning horizon and demonstrate that the cost-to-go oracle shortens the learner's planning horizon as function of its accuracy: a globally optimal oracle can shorten the planning horizon to one, leading to a one-step greedy Markov Decision Process which is much easier to optimize, while an oracle that is far away from the optimality requires planning over a longer horizon to achieve near-optimal perfor…
Hierarchical Soft Actor-Critic for Sparse-Reward Long-Horizon Reinforcement Learning
National Engineering School of Tunis, University of Tunis El Manar（突尼斯艾尔玛纳尔大学突尼斯国立工程师学校）；University of Tripoli（的黎波里大学）；University of Picardie Jules Verne Amiens（亚眠皮卡第朱尔斯·凡尔纳大学）
Seoul National University（首尔国立大学）；Artificial Intelligence Institute of Seoul National University (AIIS)（首尔国立大学人工智能研究所（AIIS））；Automation and Systems Research Institute (ASRI)（自动化与系统研究所（ASRI））
Hierarchical Reinforcement Learning (HRL) has made notable progress in complex control tasks by leveraging temporal abstraction. However, previous HRL algorithms often suffer from serious data inefficiency as environments get large. The extended components, $i.e.$, goal space and length of episodes, impose a burden on either one or both high-level and low-level policies since both levels share the total horizon of the episode. In this paper, we present a method of Decoupling Horizons Using a Graph in Hierarchical Reinforcement Learning (DHRL) which can alleviate this problem by decoupling the …
Carnegie Mellon University（卡内基梅隆大学）；Institute for Advanced Study（高等研究院）；University of California, Los Angles（加州大学洛杉矶分校）；University of Washington, Seattle（华盛顿大学西雅图分校）；Microsoft Research（微软研究院）
Learning to plan for long horizons is a central challenge in episodic reinforcement learning problems. A fundamental question is to understand how the difficulty of the problem scales as the horizon increases. Here the natural measure of sample complexity is a normalized one: we are interested in the number of episodes it takes to provably discover a policy whose value is $\varepsilon$ near to that of the optimal value, where the value is measured by the normalized cumulative reward in each episode. In a COLT 2018 open problem, Jiang and Agarwal conjectured that, for tabular, episodic reinforc…
Indian Institute of Technology (BHU)（印度理工学院（贝拿勒斯印度教大学））；Indian Institute of Technology, Hyderabad（印度理工学院海得拉巴分校）
We consider the finite horizon continuous reinforcement learning problem. Our contribution is three-fold. First,we give a tractable algorithm based on optimistic value iteration for the problem. Next,we give a lower bound on regret of order $Ω(T^{2/3})$ for any algorithm discretizes the state space, improving the previous regret bound of $Ω(T^{1/2})$ of Ortner and Ryabko \cite{contrl} for the same problem. Next,under the assumption that the rewards and transitions are Hölder Continuous we show that the upper bound on the discretization error is $const.Ln^{-α}T$. Finally,we give some simple exp…
Facebook AI Research（Facebook人工智能研究院）；CREST, ENSAE（法国国家统计与经济管理局统计研究中心, 法国国立统计与经济管理学院）；Criteo AI Lab（Criteo人工智能实验室）；Imperial College London（帝国理工学院）
Reinforcement learning algorithms are widely used in domains where it is desirable to provide a personalized service. In these domains it is common that user data contains sensitive information that needs to be protected from third parties. Motivated by this, we study privacy in the context of finite-horizon Markov Decision Processes (MDPs) by requiring information to be obfuscated on the user side. We formulate this notion of privacy for RL by leveraging the local differential privacy (LDP) framework. We establish a lower bound for regret minimization in finite-horizon MDPs with LDP guarantee…
In this paper we present Horizon, Facebook's open source applied reinforcement learning (RL) platform. Horizon is an end-to-end platform designed to solve industry applied RL problems where datasets are large (millions to billions of observations), the feedback loop is slow (vs. a simulator), and experiments must be done with care because they don't run in a simulator. Unlike other RL platforms, which are often designed for fast prototyping and experimentation, Horizon is designed with production use cases as top of mind. The platform contains workflows to train popular deep RL algorithms and …
Inverse reinforcement learning (IRL) seeks to infer a cost function that explains the underlying goals and preferences of expert demonstrations. This paper presents receding horizon inverse reinforcement learning (RHIRL), a new IRL algorithm for high-dimensional, noisy, continuous systems with black-box dynamic models. RHIRL addresses two key challenges of IRL: scalability and robustness. To handle high-dimensional continuous systems, RHIRL matches the induced optimal trajectories with expert demonstrations locally in a receding horizon manner and 'stitches' together the local solutions to lea…
Offline Reinforcement Learning with Universal Horizon Models
Interdisciplinary Program in Artificial Intelligence and ASRI, Seoul National University（人工智能交叉学科项目及首尔国立大学ASRI）；Department of Electrical and Computer Engineering, Seoul National University（电气与计算机工程系，首尔国立大学）
SPLASH! Sample-efficient Preference-based inverse reinforcement learning for Long-horizon Adversarial tasks from Suboptimal Hierarchical demonstrations
Boston University（波士顿大学）；MIT Lincoln Laboratory（麻省理工学院林肯实验室）；MIT Pavlab（麻省理工学院 Pavlab 实验室）；Woods Hole Oceanographic Institution（伍兹霍尔海洋研究所）；University of Maryland, College Park（马里兰大学学院公园分校）
In both the fields of computer science and medicine there is very strong interest in developing personalized treatment policies for patients who have variable responses to treatments. In particular, I aim to find an optimal personalized treatment policy which is a non-deterministic function of the patient specific covariate data that maximizes the expected survival time or clinical outcome. I developed an algorithmic framework to solve multistage decision problem with a varying number of stages that are subject to censoring in which the "rewards" are expected survival times. In specific, I dev…
Tsinghua University（清华大学）；University of Washington（华盛顿大学）
Episodic reinforcement learning and contextual bandits are two widely studied sequential decision-making problems. Episodic reinforcement learning generalizes contextual bandits and is often perceived to be more difficult due to long planning horizon and unknown state-dependent transitions. The current paper shows that the long planning horizon and the unknown state-dependent transitions (at most) pose little additional difficulty on sample complexity. We consider the episodic reinforcement learning with $S$ states, $A$ actions, planning horizon $H$, total reward bounded by $1$, and the agent …
We present relay policy learning, a method for imitation and reinforcement learning that can solve multi-stage, long-horizon robotic tasks. This general and universally-applicable, two-phase approach consists of an imitation learning stage that produces goal-conditioned hierarchical policies, and a reinforcement learning phase that finetunes these policies for task performance. Our method, while not necessarily perfect at imitation learning, is very amenable to further improvement via environment interaction, allowing it to scale to challenging long-horizon tasks. We simplify the long-horizon …
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Huawei Noah’s Ark Lab（华为诺亚方舟实验室）
Existing model-based value expansion methods typically leverage a world model for value estimation with a fixed rollout horizon to assist policy learning. However, the fixed rollout with an inaccurate model has a potential to harm the learning process. In this paper, we investigate the idea of using the model knowledge for value expansion adaptively. We propose a novel method called Dynamic-horizon Model-based Value Expansion (DMVE) to adjust the world model usage with different rollout horizons. Inspired by reconstruction-based techniques that can be applied for visual data novelty detection,…
Model-free reinforcement learning is known to be memory and computation efficient and more amendable to large scale problems. In this paper, two model-free algorithms are introduced for learning infinite-horizon average-reward Markov Decision Processes (MDPs). The first algorithm reduces the problem to the discounted-reward version and achieves $\mathcal{O}(T^{2/3})$ regret after $T$ steps, under the minimal assumption of weakly communicating MDPs. To our knowledge, this is the first model-free algorithm for general MDPs in this setting. The second algorithm makes use of recent advances in ada…
SINTEF Digital（SINTEF数字研究院）；Department of Engineering Cybernetics, NTNU（挪威科技大学工程控制论系）；Centre for Autonomous Marine Operations and Systems（自主海洋作业与系统中心）
Model predictive control (MPC) is a powerful trajectory optimization control technique capable of controlling complex nonlinear systems while respecting system constraints and ensuring safe operation. The MPC's capabilities come at the cost of a high online computational complexity, the requirement of an accurate model of the system dynamics, and the necessity of tuning its parameters to the specific control application. The main tunable parameter affecting the computational complexity is the prediction horizon length, controlling how far into the future the MPC predicts the system response an…
Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation
College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）；The State Key Lab of Brain-Machine Intelligence, Zhejiang University（浙江大学脑机智能国家重点实验室）；Center for AI Safety and Governance, Peking University（北京大学人工智能安全与治理中心）
Off-policy evaluation (OPE) in reinforcement learning is an important problem in settings where experimentation is limited, such as education and healthcare. But, in these very same settings, observed actions are often confounded by unobserved variables making OPE even more difficult. We study an OPE problem in an infinite-horizon, ergodic Markov decision process with unobserved confounders, where states and actions can act as proxies for the unobserved confounders. We show how, given only a latent variable model for states and actions, policy value can be identified from off-policy data. Our …
Solving hard-exploration environments in an important challenge in Reinforcement Learning. Several approaches have been proposed and studied, such as Intrinsic Motivation, co-evolution of agents and tasks, and multi-agent competition. In particular, the interplay between multiple agents has proven to be capable of generating human-relevant emergent behaviour that would be difficult or impossible to learn in single-agent settings. In this work, an extensible competitive environment for multi-agent interplay was developed, which features realistic physics and human-relevant semantics. Moreover, …
Decision-making for Autonomous Vehicles on Highway: Deep Reinforcement Learning with Continuous Action Horizon
College of Mechanical and Vehicle Engineering, Chongqing University（重庆大学机械与运载工程学院）；Department of Industrial Design, Chongqing Normal University（重庆师范大学工业设计系）
University of Alberta（阿尔伯塔大学）；Carnegie Mellon University（卡内基梅隆大学）；Google Brain（谷歌大脑）
Despite its potential to improve sample complexity versus model-free approaches, model-based reinforcement learning can fail catastrophically if the model is inaccurate. An algorithm should ideally be able to trust an imperfect model over a reasonably long planning horizon, and only rely on model-free updates when the model errors get infeasibly large. In this paper, we investigate techniques for choosing the planning horizon on a state-dependent basis, where a state's planning horizon is determined by the maximum cumulative model error around that state. We demonstrate that these state-depend…
vivo AI Lab（vivo人工智能实验室）；Zhejiang Lab（浙江实验室）；CUHK MMLab（香港大学多模态实验室）；Hubei University（湖北省大学）；Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences（中国科学院大学杭州高等研究院）
University of Toronto（多伦多大学）；Vector Institute（Vector研究所）
What goals should a multi-goal reinforcement learning agent pursue during training in long-horizon tasks? When the desired (test time) goal distribution is too distant to offer a useful learning signal, we argue that the agent should not pursue unobtainable goals. Instead, it should set its own intrinsic goals that maximize the entropy of the historical achieved goal distribution. We propose to optimize this objective by having the agent pursue past achieved goals in sparsely explored areas of the goal space, which focuses exploration on the frontier of the achievable goal set. We show that ou…
Carnegie Mellon University（卡内基梅隆大学）；University of California, Los Angles（加利福尼亚大学洛杉矶分校）
Recently there is a surge of interest in understanding the horizon-dependence of the sample complexity in reinforcement learning (RL). Notably, for an RL environment with horizon length $H$, previous work have shown that there is a probably approximately correct (PAC) algorithm that learns an $O(1)$-optimal policy using $\mathrm{polylog}(H)$ episodes of environment interactions when the number of states and actions is fixed. It is yet unknown whether the $\mathrm{polylog}(H)$ dependence is necessary or not. In this work, we resolve this question by developing an algorithm that achieves the sam…
Recent studies have shown that episodic reinforcement learning (RL) is not more difficult than contextual bandits, even with a long planning horizon and unknown state transitions. However, these results are limited to either tabular Markov decision processes (MDPs) or computationally inefficient algorithms for linear mixture MDPs. In this paper, we propose the first computationally efficient horizon-free algorithm for linear mixture MDPs, which achieves the optimal $\tilde O(d\sqrt{K} +d^2)$ regret up to logarithmic factors. Our algorithm adapts a weighted least square estimator for the unknow…
University of Torontoland（多伦多兰大学）；University of Edenborrow（伊登伯勒大学）
Off-policy evaluation of sequential decision policies from observational data is necessary in applications of batch reinforcement learning such as education and healthcare. In such settings, however, unobserved variables confound observed actions, rendering exact evaluation of new policies impossible, i.e., unidentifiable. We develop a robust approach that estimates sharp bounds on the (unidentifiable) value of a given policy in an infinite-horizon problem given data from another policy with unobserved confounding, subject to a sensitivity model. We consider stationary or baseline unobserved c…
EDF R&D（法国电力公司研发部）；FIME（FIME实验室）；University of California, Berkeley（加州大学伯克利分校）；University of Oxford（牛津大学）
We study finite-time horizon continuous-time linear-quadratic reinforcement learning problems in an episodic setting, where both the state and control coefficients are unknown to the controller. We first propose a least-squares algorithm based on continuous-time observations and controls, and establish a logarithmic regret bound of order $O((\ln M)(\ln\ln M))$, with $M$ being the number of learning episodes. The analysis consists of two parts: perturbation analysis, which exploits the regularity and robustness of the associated Riccati differential equation; and parameter estimation error, whi…
