This paper delves into designing stabilizing feedback control gains for continuous linear systems with unknown state matrix, in which the control is subject to a general structural constraint. We bring forth the ideas from reinforcement learning (RL) in conjunction with sufficient stability and performance guarantees in order to design these structured gains using the trajectory measurements of states and controls. We first formulate a model-based framework using dynamic programming (DP) to embed the structural constraint to the Linear Quadratic Regulator (LQR) gain computation in the continuous-time setting. Subsequently, we transform this LQR formulation into a policy iteration RL algorithm that can alleviate the requirement of known state matrix in conjunction with maintaining the feedback gain structure. Theoretical guarantees are provided for stability and convergence of the structured RL (SRL) algorithm. The introduced RL framework is general and can be applied to any control structure. A special control structure enabled by this RL framework is distributed learning control which is necessary for many large-scale cyber-physical systems. As such, we validate our theoretical results with numerical simulations on a multi-agent networked linear time-invariant (LTI) dynamic system.
Pacific Northwest National Laboratory (PNNL)（太平洋西北国家实验室）
This paper discusses learning a structured feedback control to obtain sufficient robustness to exogenous inputs for linear dynamic systems with unknown state matrix. The structural constraint on the controller is necessary for many cyber-physical systems, and our approach presents a design for any generic structure, paving the way for distributed learning control. The ideas from reinforcement learning (RL) in conjunction with control-theoretic sufficient stability and performance guarantees are used to develop the methodology. First, a model-based framework is formulated using dynamic programm…
Technische Universität Chemnitz（开姆尼茨工业大学）；Skolkovo Institute of Science and Technology（斯科尔科沃科学技术研究院）
Reinforcement learning is commonly associated with training of reward-maximizing (or cost-minimizing) agents, in other words, controllers. It can be applied in model-free or model-based fashion, using a priori or online collected system data to train involved parametric architectures. In general, online reinforcement learning does not guarantee closed loop stability unless special measures are taken, for instance, through learning constraints or tailored training rules. Particularly promising are hybrids of reinforcement learning with "classical" control approaches. In this work, we suggest a …
Enforcing Hard Constraints with Soft Barriers: Safe Reinforcement Learning in Unknown Stochastic Environments
Northwestern University（西北大学）；University of California, Berkeley（加州大学伯克利分校）；University of Pennsylvania（宾夕法尼亚大学）；Yale University（耶鲁大学）；University of Liverpool（利物浦大学）
Reinforcement learning is a promising paradigm for solving sequential decision-making problems, but low data efficiency and weak generalization across tasks are bottlenecks in real-world applications. Model-based meta reinforcement learning addresses these issues by learning dynamics and leveraging knowledge from prior experience. In this paper, we take a closer look at this framework, and propose a new Thompson-sampling based approach that consists of a new model to identify task dynamics together with an amortized policy optimization step. We show that our model, called a graph structured su…
We consider the problem of learning by demonstration from agents acting in unknown stochastic Markov environments or games. Our aim is to estimate agent preferences in order to construct improved policies for the same task that the agents are trying to solve. To do so, we extend previous probabilistic approaches for inverse reinforcement learning in known MDPs to the case of unknown dynamics or opponents. We do this by deriving two simplified probabilistic models of the demonstrator's policy and utility. For tractability, we use maximum a posteriori estimation rather than full Bayesian inferen…
University of Toronto（多伦多大学）；Vector Institute（向量研究所）；University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；Max Planck Institute for Software Systems (MPI-SWS)（马克斯·普朗克软件系统研究所（MPI-SWS））
We study black-box reward poisoning attacks against reinforcement learning (RL), in which an adversary aims to manipulate the rewards to mislead a sequence of RL agents with unknown algorithms to learn a nefarious policy in an environment unknown to the adversary a priori. That is, our attack makes minimum assumptions on the prior knowledge of the adversary: it has no initial knowledge of the environment or the learner, and neither does it observe the learner's internal mechanism except for its performed actions. We design a novel black-box attack, U2, that can provably achieve a near-matching…
We consider the problem of learning by demonstration from agents acting in unknown stochastic Markov environments or games. Our aim is to estimate agent preferences in order to construct improved policies for the same task that the agents are trying to solve. To do so, we extend previous probabilistic approaches for inverse reinforcement learning in known MDPs to the case of unknown dynamics or opponents. We do this by deriving two simplified probabilistic models of the demonstrator's policy and utility. For tractability, we use maximum a posteriori estimation rather than full Bayesian inferen…
Google AI（谷歌人工智能）；Technion, Israel Institute of Technology（以色列理工学院）；Tel Aviv University（特拉维夫大学）
We derive and analyze learning algorithms for apprenticeship learning, policy evaluation, and policy gradient for average reward criteria. Existing algorithms explicitly require an upper bound on the mixing time. In contrast, we build on ideas from Markov chain theory and derive sampling algorithms that do not require such an upper bound. For these algorithms, we provide theoretical bounds on their sample-complexity and running time.
Microsoft Research（微软研究院）；University of Florida（佛罗里达大学）；JD AI Research（京东人工智能研究院）
We propose a hierarchically structured reinforcement learning approach to address the challenges of planning for generating coherent multi-sentence stories for the visual storytelling task. Within our framework, the task of generating a story given a sequence of images is divided across a two-level hierarchical decoder. The high-level decoder constructs a plan by generating a semantic concept (i.e., topic) for each image in sequence. The low-level decoder generates a sentence for each image using a semantic compositional network, which effectively grounds the sentence generation conditioned on…
Equipping active colloidal robots with intelligence such that they can efficiently navigate in unknown complex environments could dramatically impact their use in emerging applications like precision surgery and targeted drug delivery. Here we develop a model-free deep reinforcement learning that can train colloidal robots to learn effective navigation strategies in unknown environments with random obstacles. We show that trained robot agents learn to make navigation decisions regarding both obstacle avoidance and travel time minimization, based solely on local sensory inputs without prior kno…
Rutgers University（罗格斯大学）；Electronics and Telecommunications Research Institute（电子与电信研究所）；Rutgers Center for Cognitive Science（罗格斯认知科学中心）
Object-centric world models provide structured representation of the scene and can be an important backbone in reinforcement learning and planning. However, existing approaches suffer in partially-observable environments due to the lack of belief states. In this paper, we propose Structured World Belief, a model for learning and inference of object-centric belief states. Inferred by Sequential Monte Carlo (SMC), our belief states provide multiple object-centric scene hypotheses. To synergize the benefits of SMC particles with object representations, we also propose a new object-centric dynamic…
University of Bristol（布里斯托大学）；CortAIx, Thales（泰雷兹公司 CortAIx）
The potential of reinforcement learning (RL) to deliver aligned and performant agents is partially bottlenecked by the reward engineering problem. One alternative to heuristic trial-and-error is preference-based RL (PbRL), where a reward function is inferred from sparse human feedback. However, prior PbRL methods lack interpretability of the learned reward structure, which hampers the ability to assess robustness and alignment. We propose an online, active preference learning algorithm that constructs reward functions with the intrinsically interpretable, compositional structure of a tree. Usi…
Indian Institute of Science（印度科学理工学院）；Indian Institute of Technology Bombay（印度理工学院孟买分校）；Xerox Research Centre India（施乐印度研究中心）
A recent goal in the Reinforcement Learning (RL) framework is to choose a sequence of actions or a policy to maximize the reward collected or minimize the regret incurred in a finite time horizon. For several RL problems in operation research and optimal control, the optimal policy of the underlying Markov Decision Process (MDP) is characterized by a known structure. The current state of the art algorithms do not utilize this known structure of the optimal policy while minimizing regret. In this work, we develop new RL algorithms that exploit the structure of the optimal policy to minimize reg…
Courant Institute of Mathematical Sciences, New York University（纽约大学库朗数学科学研究所）；New York University（纽约大学）
Reinforcement learning algorithms have had tremendous successes in online learning settings. However, these successes have relied on low-stakes interactions between the algorithmic agent and its environment. In many settings where RL could be of use, such as health care and autonomous driving, the mistakes made by most online RL algorithms during early training come with unacceptable costs. These settings require developing reinforcement learning algorithms that can operate in the so-called batch setting, where the algorithms must learn from set of data that is fixed, finite, and generated fro…
In this paper, we study the global convergence of model-based and model-free policy gradient descent and natural policy gradient descent algorithms for linear quadratic deep structured teams. In such systems, agents are partitioned into a few sub-populations wherein the agents in each sub-population are coupled in the dynamics and cost function through a set of linear regressions of the states and actions of all agents. Every agent observes its local state and the linear regressions of states, called deep states. For a sufficiently small risk factor and/or sufficiently large population, we pro…
SRT: Accelerating Reinforcement Learning via Speculative Rollout with Tree-Structured Cache
Cornell University（康奈尔大学）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Tsinghua University（清华大学）；University of Washington（华盛顿大学）；ByteDance（字节跳动）
Reinforcement Learning (RL) has been shown effective in domains where the agent can learn policies by actively interacting with its operating environment. However, if we change the RL scheme to offline setting where the agent can only update its policy via static datasets, one of the major issues in offline reinforcement learning emerges, i.e. distributional shift. We propose a Pessimistic Offline Reinforcement Learning (PessORL) algorithm to actively lead the agent back to the area where it is familiar by manipulating the value function. We focus on problems caused by out-of-distribution (OOD…
California Institute of Technology（加州理工学院）；University of California San Diego（加州大学圣地亚哥分校）；Carnegie Mellon University（卡内基梅隆大学）；Purdue University（普渡大学）
Learning a dynamical system requires stabilizing the unknown dynamics to avoid state blow-ups. However, current reinforcement learning (RL) methods lack stabilization guarantees, which limits their applicability for the control of safety-critical systems. We propose a model-based RL framework with formal stability guarantees, Krasovskii Constrained RL (KCRL), that adopts Krasovskii's family of Lyapunov functions as a stability constraint. The proposed method learns the system dynamics up to a confidence interval using feature representation, e.g. Random Fourier Features. It then solves a const…
A critical component to enabling intelligent reasoning in partially observable environments is memory. Despite this importance, Deep Reinforcement Learning (DRL) agents have so far used relatively simple memory architectures, with the main methods to overcome partial observability being either a temporal convolution over the past k frames or an LSTM layer. More recent work (Oh et al., 2016) has went beyond these architectures by using memory networks which can allow more sophisticated addressing schemes over the past k frames. But even these architectures are unsatisfactory due to the reason t…
In order to collaborate efficiently with unknown partners in cooperative control settings, adaptation of the partners based on online experience is required. The rather general and widely applicable control setting, where each cooperation partner might strive for individual goals while the control laws and objectives of the partners are unknown, entails various challenges such as the non-stationarity of the environment, the multi-agent credit assignment problem, the alter-exploration problem and the coordination problem. We propose new, modular deep decentralized Multi-Agent Reinforcement Lear…
Exploration is a fundamental challenge in reinforcement learning (RL). Many of the current exploration methods for deep RL use task-agnostic objectives, such as information gain or bonuses based on state visitation. However, many practical applications of RL involve learning more than a single task, and prior tasks can be used to inform how exploration should be performed in new tasks. In this work, we explore how prior tasks can inform an agent about how to explore effectively in new situations. We introduce a novel gradient-based fast adaptation algorithm -- model agnostic exploration with s…
Nanjing University（南京大学）；Carnegie Mellon University（卡内基梅隆大学）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Peng Cheng Laboratory（鹏城实验室）
Model-based methods have recently shown promising for offline reinforcement learning (RL), aiming to learn good policies from historical data without interacting with the environment. Previous model-based offline RL methods learn fully connected nets as world-models that map the states and actions to the next-step states. However, it is sensible that a world-model should adhere to the underlying causal effect such that it will support learning an effective policy generalizing well in unseen states. In this paper, We first provide theoretical results that causal world-models can outperform plai…
School of Computer Science and Technology, East China Normal University（华东师范大学计算机科学与技术学院）；Department of Computer Science and Engineering, Artificial Intelligence Institute, Shanghai Jiao Tong University（上海交通大学计算机科学与工程系人工智能研究所）；School of Science and Engineering, The Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳）理工学院）；School of Computational Science and Engineering, College of Computing, Georgia Institute of Technology（佐治亚理工学院计算学院计算科学与工程学院）
This work explores the large-scale multi-agent communication mechanism under a multi-agent reinforcement learning (MARL) setting. We summarize the general categories of topology for communication structures in MARL literature, which are often manually specified. Then we propose a novel framework termed as Learning Structured Communication (LSC) by using a more flexible and efficient communication topology. Our framework allows for adaptive agent grouping to form different hierarchical formations over episodes, which is generated by an auxiliary task combined with a hierarchical routing protoco…
We present a novel approach to control design for nonlinear systems which leverages model-free policy optimization techniques to learn a linearizing controller for a physical plant with unknown dynamics. Feedback linearization is a technique from nonlinear control which renders the input-output dynamics of a nonlinear plant \emph{linear} under application of an appropriate feedback controller. Once a linearizing controller has been constructed, desired output trajectories for the nonlinear plant can be tracked using a variety of linear control techniques. However, the calculation of a lineariz…
Technion – Israel Institute of Technology（以色列理工学院）；Microsoft Research（微软研究院）；Nvidia Research（英伟达研究院）
The standard feedback model of reinforcement learning requires revealing the reward of every visited state-action pair. However, in practice, it is often the case that such frequent feedback is not available. In this work, we take a first step towards relaxing this assumption and require a weaker form of feedback, which we refer to as \emph{trajectory feedback}. Instead of observing the reward obtained after every action, we assume we only receive a score that represents the quality of the whole trajectory observed by the agent, namely, the sum of all rewards obtained over this trajectory. We …
Performing autonomous exploration is essential for unmanned aerial vehicles (UAVs) operating in unknown environments. Often, these missions start with building a map for the environment via pure exploration and subsequently using (i.e. exploiting) the generated map for downstream navigation tasks. Accomplishing these navigation tasks in two separate steps is not always possible or even disadvantageous for UAVs deployed in outdoor and dynamically changing environments. Current exploration approaches either use a priori human-generated maps or use heuristics such as frontier-based exploration. O…
University of California, Berkeley（加州大学伯克利分校）；University of California, San Diego（加州大学圣地亚哥分校）；Google（谷歌）
Model-based reinforcement learning (RL) has proven to be a data efficient approach for learning control tasks but is difficult to utilize in domains with complex observations such as images. In this paper, we present a method for learning representations that are suitable for iterative model-based policy improvement, even when the underlying dynamical system has complex dynamics and image observations, in that these representations are optimized for inferring simple dynamics and cost models given data from the current policy. This enables a model-based RL method based on the linear-quadratic r…
We address reinforcement learning problems with finite state and action spaces where the underlying MDP has some known structure that could be potentially exploited to minimize the exploration rates of suboptimal (state, action) pairs. For any arbitrary structure, we derive problem-specific regret lower bounds satisfied by any learning algorithm. These lower bounds are made explicit for unstructured MDPs and for those whose transition probabilities and average reward functions are Lipschitz continuous w.r.t. the state and action. For Lipschitz MDPs, the bounds are shown not to scale with the s…
Democritus University of Thrace（色雷斯德谟克利特大学）；The Centre for Research & Technology, Hellas（希腊研究与科技中心）
This paper is an initial endeavor to bridge the gap between powerful Deep Reinforcement Learning methodologies and the problem of exploration/coverage of unknown terrains. Within this scope, MarsExplorer, an openai-gym compatible environment tailored to exploration/coverage of unknown areas, is presented. MarsExplorer translates the original robotics problem into a Reinforcement Learning setup that various off-the-shelf algorithms can tackle. Any learned policy can be straightforwardly applied to a robotic platform without an elaborate simulation model of the robot's dynamics to apply a differ…
Facebook（脸书）；Lamsade, Univ. Paris Dauphine（巴黎多芬纳大学Lamsade实验室）
Effective coordination is crucial to solve multi-agent collaborative (MAC) problems. While centralized reinforcement learning methods can optimally solve small MAC instances, they do not scale to large problems and they fail to generalize to scenarios different from those seen during training. In this paper, we consider MAC problems with some intrinsic notion of locality (e.g., geographic proximity) such that interactions between agents and tasks are locally limited. By leveraging this property, we introduce a novel structured prediction approach to assign agents to tasks. At each step, the as…
TemplateRL: Structured Template-Guided Reinforcement Learning for LLM Reasoning
Tsinghua University（清华大学）；Nanyang Technological University（南洋理工大学）；Princeton University（普林斯顿大学）；Shanghai AI Lab（上海人工智能实验室）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）
Department of XXX, University of YYY, Location, Country；School of ZZZ, Institute of WWW, Location, Country
Assembly of multi-part physical structures is both a valuable end product for autonomous robotics, as well as a valuable diagnostic task for open-ended training of embodied intelligent agents. We introduce a naturalistic physics-based environment with a set of connectable magnet blocks inspired by children's toy kits. The objective is to assemble blocks into a succession of target blueprints. Despite the simplicity of this objective, the compositional nature of building diverse blueprints from a set of blocks leads to an explosion of complexity in structures that agents encounter. Furthermore,…
Reinforcement learning would enjoy better success on real-world problems if domain knowledge could be imparted to the algorithm by the modelers. Most problems have both hidden state and unknown dynamics. Partially observable Markov decision processes (POMDPs) allow for the modeling of both. Unfortunately, they do not provide a natural framework in which to specify knowledge about the domain dynamics. The designer must either admit to knowing nothing about the dynamics or completely specify the dynamics (thereby turning it into a planning problem). We propose a new framework called a partially …
Estimating unknown parameters in differential equations with a reinforcement learning based PSO method
School of Software Technology, Dalian University of Technology（大连理工大学软件学院）；Beijing Institute of Mathematical Sciences and Applications（北京数学科学与应用研究院）；Meinig School of Biomedical Engineering, Cornell University（康奈尔大学迈宁生物医学工程学院）；Yau Mathematical Sciences Center, Tsinghua University（清华大学丘成桐数学科学中心）
