Reinforcement learning algorithms are typically designed to learn a performant policy that can repeatedly and autonomously complete a task, usually starting from scratch. However, in many real-world situations, the goal might not be to learn a policy that can do the task repeatedly, but simply to perform a new task successfully once in a single trial. For example, imagine a disaster relief robot tasked with retrieving an item from a fallen building, where it cannot get direct supervision from humans. It must retrieve this object within one test-time trial, and must do so while tackling unknown obstacles, though it may leverage knowledge it has of the building before the disaster. We formalize this problem setting, which we call single-life reinforcement learning (SLRL), where an agent must complete a task within a single episode without interventions, utilizing its prior experience while contending with some form of novelty. SLRL provides a natural setting to study the challenge of autonomously adapting to unfamiliar situations, and we find that algorithms designed for standard episodic reinforcement learning often struggle to recover from out-of-distribution states in this setting. Motivated by this observation, we propose an algorithm, $Q$-weighted adversarial learning (QWALE), which employs a distribution matching strategy that leverages the agent's prior experience as guidance in novel situations. Our experiments on several single-life continuous control problems indicate that methods based on our distribution matching formulation are 20-60% more successful because they can more quickly recover from novel states.
Providing densely shaped reward functions for RL algorithms is often exceedingly challenging, motivating the development of RL algorithms that can learn from easier-to-specify sparse reward functions. This sparsity poses new exploration challenges. One common way to address this problem is using demonstrations to provide initial signal about regions of the state space with high rewards. However, prior RL from demonstrations algorithms introduce significant complexity and many hyperparameters, making them hard to implement and tune. We introduce Monte Carlo Augmented Actor Critic (MCAC), a parameter free modification to standard actor-critic algorithms which initializes the replay buffer with demonstrations and computes a modified $Q$-value by taking the maximum of the standard temporal distance (TD) target and a Monte Carlo estimate of the reward-to-go. This encourages exploration in the neighborhood of high-performing trajectories by encouraging high $Q$-values in corresponding regions of the state space. Experiments across $5$ continuous control domains suggest that MCAC can be used to significantly increase learning efficiency across $6$ commonly used RL and RL-from-demonstrations algorithms. See https://sites.google.com/view/mcac-rl for code and supplementary material.
In offline model-based reinforcement learning (offline MBRL), we learn a dynamic model from historically collected data, and subsequently utilize the learned model and fixed datasets for policy learning, without further interacting with the environment. Offline MBRL algorithms can improve the efficiency and stability of policy learning over the model-free algorithms. However, in most of the existing offline MBRL algorithms, the learning objectives for the dynamic models and the policies are isolated from each other. Such an objective mismatch may lead to inferior performance of the learned agents. In this paper, we address this issue by developing an iterative offline MBRL framework, where we maximize a lower bound of the true expected return, by alternating between dynamic-model training and policy learning. With the proposed unified model-policy learning framework, we achieve competitive performance on a wide range of continuous-control offline reinforcement learning datasets. Source code is publicly released.
We consider the problem of learning control policies in discrete-time stochastic systems which guarantee that the system stabilizes within some specified stabilization region with probability~$1$. Our approach is based on the novel notion of stabilizing ranking supermartingales (sRSMs) that we introduce in this work. Our sRSMs overcome the limitation of methods proposed in previous works whose applicability is restricted to systems in which the stabilizing region cannot be left once entered under any control policy. We present a learning procedure that learns a control policy together with an sRSM that formally certifies probability~$1$ stability, both learned as neural networks. We show that this procedure can also be adapted to formally verifying that, under a given Lipschitz continuous control policy, the stochastic system stabilizes within some stabilizing region with probability~$1$. Our experimental evaluation shows that our learning procedure can successfully learn provably stabilizing policies in practice.
Offline reinforcement learning (RL) addresses the problem of learning a performant policy from a fixed batch of data collected by following some behavior policy. Model-based approaches are particularly appealing in the offline setting since they can extract more learning signals from the logged dataset by learning a model of the environment. However, the performance of existing model-based approaches falls short of model-free counterparts, due to the compounding of estimation errors in the learned model. Driven by this observation, we argue that it is critical for a model-based method to understand when to trust the model and when to rely on model-free estimates, and how to act conservatively w.r.t. both. To this end, we derive an elegant and simple methodology called conservative Bayesian model-based value expansion for offline policy optimization (CBOP), that trades off model-free and model-based estimates during the policy evaluation step according to their epistemic uncertainties, and facilitates conservatism by taking a lower bound on the Bayesian posterior value estimate. On the standard D4RL continuous control tasks, we find that our method significantly outperforms previous model-based approaches: e.g., MOPO by $116.4$%, MOReL by $23.2$% and COMBO by $23.7$%. Further, CBOP achieves state-of-the-art performance on $11$ out of $18$ benchmark datasets while doing on par on the remaining datasets.
High variances in reinforcement learning have shown impeding successful convergence and hurting task performance. As reward signal plays an important role in learning behavior, multi-step methods have been considered to mitigate the problem, and are believed to be more effective than single step methods. However, there is a lack of comprehensive and systematic study on this important aspect to demonstrate the effectiveness of multi-step methods in solving highly complex continuous control problems. In this study, we introduce a new long $N$-step surrogate stage (LNSS) reward approach to effectively account for complex environment dynamics while previous methods are usually feasible for limited number of steps. The LNSS method is simple, low computational cost, and applicable to value based or policy gradient reinforcement learning. We systematically evaluate LNSS in OpenAI Gym and DeepMind Control Suite to address some complex benchmark environments that have been challenging to obtain good results by DRL in general. We demonstrate performance improvement in terms of total reward, convergence speed, and coefficient of variation (CV) by LNSS. We also provide analytical insights on how LNSS exponentially reduces the upper bound on the variances of Q value from a respective single step method
Gradient-based methods have been widely used for system design and optimization in diverse application domains. Recently, there has been a renewed interest in studying theoretical properties of these methods in the context of control and reinforcement learning. This article surveys some of the recent developments on policy optimization, a gradient-based iterative approach for feedback control synthesis, popularized by successes of reinforcement learning. We take an interdisciplinary perspective in our exposition that connects control theory, reinforcement learning, and large-scale optimization. We review a number of recently-developed theoretical results on the optimization landscape, global convergence, and sample complexity of gradient-based methods for various continuous control problems such as the linear quadratic regulator (LQR), $\mathcal{H}_\infty$ control, risk-sensitive control, linear quadratic Gaussian (LQG) control, and output feedback synthesis. In conjunction with these optimization results, we also discuss how direct policy optimization handles stability and robustness concerns in learning-based control, two main desiderata in control engineering. We conclude the survey by pointing out several challenges and opportunities at the intersection of learning and control.
We consider how to most efficiently leverage teleoperator time to collect data for learning robust image-based value functions and policies for sparse reward robotic tasks. To accomplish this goal, we modify the process of data collection to include more than just successful demonstrations of the desired task. Instead we develop a novel protocol that we call Visual Backtracking Teleoperation (VBT), which deliberately collects a dataset of visually similar failures, recoveries, and successes. VBT data collection is particularly useful for efficiently learning accurate value functions from small datasets of image-based observations. We demonstrate VBT on a real robot to perform continuous control from image observations for the deformable manipulation task of T-shirt grasping. We find that by adjusting the data collection process we improve the quality of both the learned value functions and policies over a variety of baseline methods for data collection. Specifically, we find that offline reinforcement learning on VBT data outperforms standard behavior cloning on successful demonstration data by 13% when both methods are given equal-sized datasets of 60 minutes of data from the real robot.
Real-world deployment of machine learning models is challenging because data evolves over time. While no model can work when data evolves in an arbitrary fashion, if there is some pattern to these changes, we might be able to design methods to address it. This paper addresses situations when data evolves gradually. We introduce a time-varying propensity score that can detect gradual shifts in the distribution of data which allows us to selectively sample past data to update the model -- not just similar data from the past like that of a standard propensity score but also data that evolved in a similar fashion in the past. The time-varying propensity score is quite general: we demonstrate different ways of implementing it and evaluate it on a variety of problems ranging from supervised learning (e.g., image classification problems) where data undergoes a sequence of gradual shifts, to reinforcement learning tasks (e.g., robotic manipulation and continuous control) where data shifts as the policy or the task changes.
This paper explores continuous-time control synthesis for target-driven navigation to satisfy complex high-level tasks expressed as linear temporal logic (LTL). We propose a model-free framework using deep reinforcement learning (DRL) where the underlying dynamic system is unknown (an opaque box). Unlike prior work, this paper considers scenarios where the given LTL specification might be infeasible and therefore cannot be accomplished globally. Instead of modifying the given LTL formula, we provide a general DRL-based approach to satisfy it with minimal violation. To do this, we transform a previously multi-objective DRL problem, which requires simultaneous automata satisfaction and minimum violation cost, into a single objective. By guiding the DRL agent with a sampling-based path planning algorithm for the potentially infeasible LTL task, the proposed approach mitigates the myopic tendencies of DRL, which are often an issue when learning general LTL tasks that can have long or infinite horizons. This is achieved by decomposing an infeasible LTL formula into several reach-avoid sub-tasks with shorter horizons, which can be trained in a modular DRL architecture. Furthermore, we overcome the challenge of the exploration process for DRL in complex and cluttered environments by using path planners to design rewards that are dense in the configuration space. The benefits of the presented approach are demonstrated through testing on various complex nonlinear systems and compared with state-of-the-art baselines. The Video demonstration can be found here:https://youtu.be/jBhx6Nv224E.
Online planning is crucial for high performance in many complex sequential decision-making tasks. Monte Carlo Tree Search (MCTS) employs a principled mechanism for trading off exploration for exploitation for efficient online planning, and it outperforms comparison methods in many discrete decision-making domains such as Go, Chess, and Shogi. Subsequently, extensions of MCTS to continuous domains have been developed. However, the inherent high branching factor and the resulting explosion of the search tree size are limiting the existing methods. To address this problem, we propose Continuous Monte Carlo Graph Search (CMCGS), an extension of MCTS to online planning in environments with continuous state and action spaces. CMCGS takes advantage of the insight that, during planning, sharing the same action policy between several states can yield high performance. To implement this idea, at each time step, CMCGS clusters similar states into a limited number of stochastic action bandit nodes, which produce a layered directed graph instead of an MCTS search tree. Experimental evaluation shows that CMCGS outperforms comparable planning methods in several complex continuous DeepMind Control Suite benchmarks and 2D navigation and exploration tasks with limited sample budgets. Furthermore, CMCGS can be scaled up through parallelization, and it outperforms the Cross-Entropy Method (CEM) in continuous control with learned dynamics models.
Actor-critic (AC) algorithms are a class of model-free deep reinforcement learning algorithms, which have proven their efficacy in diverse domains, especially in solving continuous control problems. Improvement of exploration (action entropy) and exploitation (expected return) using more efficient samples is a critical issue in AC algorithms. A basic strategy of a learning algorithm is to facilitate indiscriminately exploring all of the environment state space, as well as to encourage exploring rarely visited states rather than frequently visited one. Under this strategy, we propose a new method to boost exploration through an intrinsic reward, based on measurement of a state's novelty and the associated benefit of exploring the state (with regards to policy optimization), altogether called plausible novelty. With incentivized exploration of plausible novel states, an AC algorithm is able to improve its sample efficiency and hence training performance. The new method is verified by extensive simulations of continuous control tasks of MuJoCo environments on a variety of prominent off-policy AC algorithms.
While the maximum entropy (MaxEnt) reinforcement learning (RL) framework -- often touted for its exploration and robustness capabilities -- is usually motivated from a probabilistic perspective, the use of deep probabilistic models has not gained much traction in practice due to their inherent complexity. In this work, we propose the adoption of latent variable policies within the MaxEnt framework, which we show can provably approximate any policy distribution, and additionally, naturally emerges under the use of world models with a latent belief state. We discuss why latent variable policies are difficult to train, how naive approaches can fail, then subsequently introduce a series of improvements centered around low-cost marginalization of the latent state, allowing us to make full use of the latent state at minimal additional cost. We instantiate our method under the actor-critic framework, marginalizing both the actor and critic. The resulting algorithm, referred to as Stochastic Marginal Actor-Critic (SMAC), is simple yet effective. We experimentally validate our method on continuous control tasks, showing that effective marginalization can lead to better exploration and more robust training. Our implementation is open sourced at https://github.com/zdhNarsil/Stochastic-Marginal-Actor-Critic.
In continuous control, exploration is often performed through undirected strategies in which parameters of the networks or selected actions are perturbed by random noise. Although the deep setting of undirected exploration has been shown to improve the performance of on-policy methods, they introduce an excessive computational complexity and are known to fail in the off-policy setting. The intrinsically motivated exploration is an effective alternative to the undirected strategies, but they are usually studied for discrete action domains. In this paper, we investigate how intrinsic motivation can effectively be combined with deep reinforcement learning in the control of continuous systems to obtain a directed exploratory behavior. We adapt the existing theories on animal motivational systems into the reinforcement learning paradigm and introduce a novel and scalable directed exploration strategy. The introduced approach, motivated by the maximization of the value function's error, can benefit from a collected set of experiences by extracting useful information and unify the intrinsic exploration motivations in the literature under a single exploration objective. An extensive set of empirical studies demonstrate that our framework extends to larger and more diverse state spaces, dramatically improves the baselines, and outperforms the undirected strategies significantly.
Switched-actuator systems with setup times (SAcSSs) are systems in which the actuator configuration has to be switched during operation, and where the switching induces non-negligible actuator downtime. Optimally controlling SAcSSs requires the online solving of both a discrete actuator allocation problem, in which the switch-induced actuator downtime is taken into account, as well as an optimization problem for the (typically continuous) control inputs. Mixed-integer model predictive control (MI-MPC) offers a powerful framework for tackling such problems. However, the efficient modeling of SAcSSs for MI-MPC is not straightforward, and real-time feasibility is often a major hurdle in practice. It is the objective of this paper to provide an intuitive and systematic modeling procedure tailored to SAcSSs, which is specifically designed to allow for user-friendly controller synthesis, and to yield efficient MI-MPCs. We apply these new results in a case study of large-volume magnetic-resonance-guided high-intensity focused ultrasound hyperthermia, which involves the heating of tumors (using real-valued local heating controls, as well as discrete range-extending actuator relocation during which no heating is allowed) to enhance the efficacy of radio- and chemotherapy.
In reinforcement learning for safety-critical settings, it is often desirable for the agent to obey safety constraints at all points in time, including during training. We present a novel neurosymbolic approach called SPICE to solve this safe exploration problem. SPICE uses an online shielding layer based on symbolic weakest preconditions to achieve a more precise safety analysis than existing tools without unduly impacting the training process. We evaluate the approach on a suite of continuous control benchmarks and show that it can achieve comparable performance to existing safe learning techniques while incurring fewer safety violations. Additionally, we present theoretical results showing that SPICE converges to the optimal safe policy under reasonable assumptions.
The ability of continual learning systems to transfer knowledge from previously seen tasks in order to maximize performance on new tasks is a significant challenge for the field, limiting the applicability of continual learning solutions to realistic scenarios. Consequently, this study aims to broaden our understanding of transfer and its driving forces in the specific case of continual reinforcement learning. We adopt SAC as the underlying RL algorithm and Continual World as a suite of continuous control tasks. We systematically study how different components of SAC (the actor and the critic, exploration, and data) affect transfer efficacy, and we provide recommendations regarding various modeling options. The best set of choices, dubbed ClonEx-SAC, is evaluated on the recent Continual World benchmark. ClonEx-SAC achieves 87% final success rate compared to 80% of PackNet, the best method in the benchmark. Moreover, the transfer grows from 0.18 to 0.54 according to the metric provided by Continual World.
Biological and artificial agents need to deal with constant changes in the real world. We study this problem in four classical continuous control environments, augmented with morphological perturbations. Learning to locomote when the length and the thickness of different body parts vary is challenging, as the control policy is required to adapt to the morphology to successfully balance and advance the agent. We show that a control policy based on the proprioceptive state performs poorly with highly variable body configurations, while an (oracle) agent with access to a learned encoding of the perturbation performs significantly better. We introduce DMAP, a biologically-inspired, attention-based policy network architecture. DMAP combines independent proprioceptive processing, a distributed policy with individual controllers for each joint, and an attention mechanism, to dynamically gate sensory information from different body parts to different controllers. Despite not having access to the (hidden) morphology information, DMAP can be trained end-to-end in all the considered environments, overall matching or surpassing the performance of an oracle agent. Thus DMAP, implementing principles from biological motor control, provides a strong inductive bias for learning challenging sensorimotor tasks. Overall, our work corroborates the power of these principles in challenging locomotion tasks.
The growing complexity and capacity demands for mobile networks necessitate innovative techniques for optimizing resource usage. Meanwhile, recent breakthroughs have brought Reinforcement Learning (RL) into the domain of continuous control of real-world systems. As a step towards RL-based network control, this paper introduces a new framework for benchmarking the performance of an RL agent in network environments simulated with ns-3. Within this framework, we demonstrate that an RL agent without domain-specific knowledge can learn how to efficiently adjust Radio Access Network (RAN) parameters to match offline optimization in static scenarios, while also adapting on the fly in dynamic scenarios, in order to improve the overall user experience. Our proposed framework may serve as a foundation for further work in developing workflows for designing RL-based RAN control algorithms.
Existing imitation learning methods mainly focus on making an agent effectively mimic a demonstrated behavior, but do not address the potential contradiction between the behavior style and the objective of a task. There is a general lack of efficient methods that allow an agent to partially imitate a demonstrated behavior to varying degrees, while completing the main objective of a task. In this paper we propose a method called Regularized Soft Actor-Critic which formulates the main task and the imitation task under the Constrained Markov Decision Process framework (CMDP). The main task is defined as the maximum entropy objective used in Soft Actor-Critic (SAC) and the imitation task is defined as a constraint. We evaluate our method on continuous control tasks relevant to video games applications.