PyTorchKR Papers

최신 논문

총 300편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2404.09080

Safe Reinforcement Learning on the Constraint Manifold: Theory and Applications

Integrating learning-based techniques, especially reinforcement learning, into robotics is promising for solving complex problems in unstructured environments. However, most existing approaches are trained in well-tuned simulators and subsequently deployed on real robots without online fine-tuning. In this setting, extensive engineering is required to mitigate the sim-to-real gap, which can be challenging for complex systems. Instead, learning with real-world interaction data offers a promising alternative: it not only eliminates the need for a fine-tuned simulator but also applies to a broader range of tasks where accurate modeling is unfeasible. One major problem for on-robot reinforcement learning is ensuring safety, as uncontrolled exploration can cause catastrophic damage to the robot or the environment. Indeed, safety specifications, often represented as constraints, can be complex and non-linear, making safety challenging to guarantee in learning systems. In this paper, we show how we can impose complex safety constraints on learning-based robotics systems in a principled manner, both from theoretical and practical points of view. Our approach is based on the concept of the Constraint Manifold, representing the set of safe robot configurations. Exploiting differential geometry techniques, i.e., the tangent space, we can construct a safe action space, allowing learning agents to sample arbitrary actions while ensuring safety. We demonstrate the method's effectiveness in a real-world Robot Air Hockey task, showing that our method can handle high-dimensional tasks with complex constraints. Videos of the real robot experiments are available on the project website (https://puzeliu.github.io/TRO-ATACOM).

arXiv:2403.18209

Long and Short-Term Constraints Driven Safe Reinforcement Learning for Autonomous Driving

Reinforcement learning (RL) has been widely used in decision-making and control tasks, but the risk is very high for the agent in the training process due to the requirements of interaction with the environment, which seriously limits its industrial applications such as autonomous driving systems. Safe RL methods are developed to handle this issue by constraining the expected safety violation costs as a training objective, but the occurring probability of an unsafe state is still high, which is unacceptable in autonomous driving tasks. Moreover, these methods are difficult to achieve a balance between the cost and return expectations, which leads to learning performance degradation for the algorithms. In this paper, we propose a novel algorithm based on the long and short-term constraints (LSTC) for safe RL. The short-term constraint aims to enhance the short-term state safety that the vehicle explores, while the long-term constraint enhances the overall safety of the vehicle throughout the decision-making process, both of which are jointly used to enhance the vehicle safety in the training process. In addition, we develop a safe RL method with dual-constraint optimization based on the Lagrange multiplier to optimize the training process for end-to-end autonomous driving. Comprehensive experiments were conducted on the MetaDrive simulator. Experimental results demonstrate that the proposed method achieves higher safety in continuous state and action tasks, and exhibits higher exploration performance in long-distance decision-making tasks compared with state-of-the-art methods.

arXiv:2403.15928

Safe Reinforcement Learning for Constrained Markov Decision Processes with Stochastic Stopping Time

In this paper, we present an online reinforcement learning algorithm for constrained Markov decision processes with a safety constraint. Despite the necessary attention of the scientific community, considering stochastic stopping time, the problem of learning optimal policy without violating safety constraints during the learning phase is yet to be addressed. To this end, we propose an algorithm based on linear programming that does not require a process model. We show that the learned policy is safe with high confidence. We also propose a method to compute a safe baseline policy, which is central in developing algorithms that do not violate the safety constraints. Finally, we provide simulation results to show the efficacy of the proposed algorithm. Further, we demonstrate that efficient exploration can be achieved by defining a subset of the state-space called proxy set.

arXiv:2403.13213공식 코드 · Papers with Code

From Representational Harms to Quality-of-Service Harms: A Case Study on Llama 2 Safety Safeguards

Recent progress in large language models (LLMs) has led to their widespread adoption in various domains. However, these advancements have also introduced additional safety risks and raised concerns regarding their detrimental impact on already marginalized populations. Despite growing mitigation efforts to develop safety safeguards, such as supervised safety-oriented fine-tuning and leveraging safe reinforcement learning from human feedback, multiple concerns regarding the safety and ingrained biases in these models remain. Furthermore, previous work has demonstrated that models optimized for safety often display exaggerated safety behaviors, such as a tendency to refrain from responding to certain requests as a precautionary measure. As such, a clear trade-off between the helpfulness and safety of these models has been documented in the literature. In this paper, we further investigate the effectiveness of safety measures by evaluating models on already mitigated biases. Using the case of Llama 2 as an example, we illustrate how LLMs' safety responses can still encode harmful assumptions. To do so, we create a set of non-toxic prompts, which we then use to evaluate Llama models. Through our new taxonomy of LLMs responses to users, we observe that the safety/helpfulness trade-offs are more pronounced for certain demographic groups which can lead to quality-of-service harms for marginalized populations.

arXiv:2403.12847공식 코드 · Papers with Code

Policy Bifurcation in Safe Reinforcement Learning

Safe reinforcement learning (RL) offers advanced solutions to constrained optimal control problems. Existing studies in safe RL implicitly assume continuity in policy functions, where policies map states to actions in a smooth, uninterrupted manner; however, our research finds that in some scenarios, the feasible policy should be discontinuous or multi-valued, interpolating between discontinuous local optima can inevitably lead to constraint violations. We are the first to identify the generating mechanism of such a phenomenon, and employ topological analysis to rigorously prove the existence of policy bifurcation in safe RL, which corresponds to the contractibility of the reachable tuple. Our theorem reveals that in scenarios where the obstacle-free state space is non-simply connected, a feasible policy is required to be bifurcated, meaning its output action needs to change abruptly in response to the varying state. To train such a bifurcated policy, we propose a safe RL algorithm called multimodal policy optimization (MUPO), which utilizes a Gaussian mixture distribution as the policy output. The bifurcated behavior can be achieved by selecting the Gaussian component with the highest mixing coefficient. Besides, MUPO also integrates spectral normalization and forward KL divergence to enhance the policy's capability of exploring different modes. Experiments with vehicle control tasks show that our algorithm successfully learns the bifurcated policy and ensures satisfying safety, while a continuous policy suffers from inevitable constraint violations.

arXiv:2403.04007공식 코드 · Papers with Code

Sampling-based Safe Reinforcement Learning for Nonlinear Dynamical Systems

We develop provably safe and convergent reinforcement learning (RL) algorithms for control of nonlinear dynamical systems, bridging the gap between the hard safety guarantees of control theory and the convergence guarantees of RL theory. Recent advances at the intersection of control and RL follow a two-stage, safety filter approach to enforcing hard safety constraints: model-free RL is used to learn a potentially unsafe controller, whose actions are projected onto safe sets prescribed, for example, by a control barrier function. Though safe, such approaches lose any convergence guarantees enjoyed by the underlying RL methods. In this paper, we develop a single-stage, sampling-based approach to hard constraint satisfaction that learns RL controllers enjoying classical convergence guarantees while satisfying hard safety constraints throughout training and deployment. We validate the efficacy of our approach in simulation, including safe control of a quadcopter in a challenging obstacle avoidance problem, and demonstrate that it outperforms existing benchmarks.

arXiv:2312.09468

Safe Reinforcement Learning in a Simulated Robotic Arm

Reinforcement learning (RL) agents need to explore their environments in order to learn optimal policies. In many environments and tasks, safety is of critical importance. The widespread use of simulators offers a number of advantages, including safe exploration which will be inevitable in cases when RL systems need to be trained directly in the physical environment (e.g. in human-robot interaction). The popular Safety Gym library offers three mobile agent types that can learn goal-directed tasks while considering various safety constraints. In this paper, we extend the applicability of safe RL algorithms by creating a customized environment with Panda robotic arm where Safety Gym algorithms can be tested. We performed pilot experiments with the popular PPO algorithm comparing the baseline with the constrained version and show that the constrained version is able to learn the equally good policy while better complying with safety constraints and taking longer training time as expected.

arXiv:2402.17217

Temporal Logic Specification-Conditioned Decision Transformer for Offline Safe Reinforcement Learning

Offline safe reinforcement learning (RL) aims to train a constraint satisfaction policy from a fixed dataset. Current state-of-the-art approaches are based on supervised learning with a conditioned policy. However, these approaches fall short in real-world applications that involve complex tasks with rich temporal and logical structures. In this paper, we propose temporal logic Specification-conditioned Decision Transformer (SDT), a novel framework that harnesses the expressive power of signal temporal logic (STL) to specify complex temporal rules that an agent should follow and the sequential modeling capability of Decision Transformer (DT). Empirical evaluations on the DSRL benchmarks demonstrate the better capacity of SDT in learning safe and high-reward policies compared with existing approaches. In addition, SDT shows good alignment with respect to different desired degrees of satisfaction of the STL specification that it is conditioned on.

arXiv:2402.15650

Uniformly Safe RL with Objective Suppression for Multi-Constraint Safety-Critical Applications

Safe reinforcement learning tasks are a challenging domain despite being very common in the real world. The widely adopted CMDP model constrains the risks in expectation, which makes room for dangerous behaviors in long-tail states. In safety-critical domains, such behaviors could lead to disastrous outcomes. To address this issue, we first describe the problem with a stronger Uniformly Constrained MDP (UCMDP) model where we impose constraints on all reachable states; we then propose Objective Suppression, a novel method that adaptively suppresses the task reward maximizing objectives according to a safety critic, as a solution to the Lagrangian dual of a UCMDP. We benchmark Objective Suppression in two multi-constraint safety domains, including an autonomous driving domain where any incorrect behavior can lead to disastrous consequences. On the driving domain, we evaluate on open source and proprietary data and evaluate transfer to a real autonomous fleet. Empirically, we demonstrate that our proposed method, when combined with existing safe RL algorithms, can match the task reward achieved by baselines with significantly fewer constraint violations.

arXiv:2402.15197

Safety Optimized Reinforcement Learning via Multi-Objective Policy Optimization

Safe reinforcement learning (Safe RL) refers to a class of techniques that aim to prevent RL algorithms from violating constraints in the process of decision-making and exploration during trial and error. In this paper, a novel model-free Safe RL algorithm, formulated based on the multi-objective policy optimization framework is introduced where the policy is optimized towards optimality and safety, simultaneously. The optimality is achieved by the environment reward function that is subsequently shaped using a safety critic. The advantage of the Safety Optimized RL (SORL) algorithm compared to the traditional Safe RL algorithms is that it omits the need to constrain the policy search space. This allows SORL to find a natural tradeoff between safety and optimality without compromising the performance in terms of either safety or optimality due to strict search space constraints. Through our theoretical analysis of SORL, we propose a condition for SORL's converged policy to guarantee safety and then use it to introduce an aggressiveness parameter that allows for fine-tuning the mentioned tradeoff. The experimental results obtained in seven different robotic environments indicate a considerable reduction in the number of safety violations along with higher, or competitive, policy returns, in comparison to six different state-of-the-art Safe RL methods. The results demonstrate the significant superiority of the proposed SORL algorithm in safety-critical applications.

arXiv:2402.08502

Provable Traffic Rule Compliance in Safe Reinforcement Learning on the Open Sea

For safe operation, autonomous vehicles have to obey traffic rules that are set forth in legal documents formulated in natural language. Temporal logic is a suitable concept to formalize such traffic rules. Still, temporal logic rules often result in constraints that are hard to solve using optimization-based motion planners. Reinforcement learning (RL) is a promising method to find motion plans for autonomous vehicles. However, vanilla RL algorithms are based on random exploration and do not automatically comply with traffic rules. Our approach accomplishes guaranteed rule-compliance by integrating temporal logic specifications into RL. Specifically, we consider the application of vessels on the open sea, which must adhere to the Convention on the International Regulations for Preventing Collisions at Sea (COLREGS). To efficiently synthesize rule-compliant actions, we combine predicates based on set-based prediction with a statechart representing our formalized rules and their priorities. Action masking then restricts the RL agent to this set of verified rule-compliant actions. In numerical evaluations on critical maritime traffic situations, our agent always complies with the formalized legal rules and never collides while achieving a high goal-reaching rate during training and deployment. In contrast, vanilla and traffic rule-informed RL agents frequently violate traffic rules and collide even after training.

arXiv:2402.05412

Techno-Economic Modeling and Safe Operational Optimization of Multi-Network Constrained Integrated Community Energy Systems

The integrated community energy system (ICES) has emerged as a promising solution for enhancing the efficiency of the distribution system by effectively coordinating multiple energy sources. However, the operational optimization of ICES is hindered by the physical constraints of heterogeneous networks including electricity, natural gas, and heat. These challenges are difficult to address due to the non-linearity of network constraints and the high complexity of multi-network coordination. This paper, therefore, proposes a novel Safe Reinforcement Learning (SRL) algorithm to optimize the multi-network constrained operation problem of ICES. Firstly, a comprehensive ICES model is established considering integrated demand response (IDR), multiple energy devices, and network constraints. The multi-network operational optimization problem of ICES is then presented and reformulated as a constrained Markov Decision Process (C-MDP) accounting for violating physical network constraints. The proposed novel SRL algorithm, named Primal-Dual Twin Delayed Deep Deterministic Policy Gradient (PD-TD3), solves the C-MDP by employing a Lagrangian multiplier to penalize the multi-network constraint violation, ensuring that violations are within a tolerated range and avoid over-conservative strategy with a low reward at the same time. The proposed algorithm accurately estimates the cumulative reward and cost of the training process, thus achieving a fair balance between improving profits and reducing constraint violations in a privacy-protected environment with only partial information. A case study comparing the proposed algorithm with benchmark RL algorithms demonstrates the computational performance in increasing total profits and alleviating the network constraint violations.

arXiv:2402.02025

A Survey of Constraint Formulations in Safe Reinforcement Learning

Safety is critical when applying reinforcement learning (RL) to real-world problems. As a result, safe RL has emerged as a fundamental and powerful paradigm for optimizing an agent's policy while incorporating notions of safety. A prevalent safe RL approach is based on a constrained criterion, which seeks to maximize the expected cumulative reward subject to specific safety constraints. Despite recent effort to enhance safety in RL, a systematic understanding of the field remains difficult. This challenge stems from the diversity of constraint representations and little exploration of their interrelations. To bridge this knowledge gap, we present a comprehensive review of representative constraint formulations, along with a curated selection of algorithms designed specifically for each formulation. In addition, we elucidate the theoretical underpinnings that reveal the mathematical mutual relations among common problem formulations. We conclude with a discussion of the current state and future directions of safe reinforcement learning research.

arXiv:2402.02624

A Safe Reinforcement Learning driven Weights-varying Model Predictive Control for Autonomous Vehicle Motion Control

Determining the optimal cost function parameters of Model Predictive Control (MPC) to optimize multiple control objectives is a challenging and time-consuming task. Multiobjective Bayesian Optimization (BO) techniques solve this problem by determining a Pareto optimal parameter set for an MPC with static weights. However, a single parameter set may not deliver the most optimal closed-loop control performance when the context of the MPC operating conditions changes during its operation, urging the need to adapt the cost function weights at runtime. Deep Reinforcement Learning (RL) algorithms can automatically learn context-dependent optimal parameter sets and dynamically adapt for a Weightsvarying MPC (WMPC). However, learning cost function weights from scratch in a continuous action space may lead to unsafe operating states. To solve this, we propose a novel approach limiting the RL actions within a safe learning space representing a catalog of pre-optimized BO Pareto-optimal weight sets. We conceive a RL agent not to learn in a continuous space but to proactively anticipate upcoming control tasks and to choose the most optimal discrete actions, each corresponding to a single set of Pareto optimal weights, context-dependent. Hence, even an untrained RL agent guarantees a safe and optimal performance. Experimental results demonstrate that an untrained RL-WMPC shows Pareto-optimal closed-loop behavior and training the RL-WMPC helps exhibit a performance beyond the Pareto-front.

arXiv:2402.00816

Leveraging Approximate Model-based Shielding for Probabilistic Safety Guarantees in Continuous Environments

Shielding is a popular technique for achieving safe reinforcement learning (RL). However, classical shielding approaches come with quite restrictive assumptions making them difficult to deploy in complex environments, particularly those with continuous state or action spaces. In this paper we extend the more versatile approximate model-based shielding (AMBS) framework to the continuous setting. In particular we use Safety Gym as our test-bed, allowing for a more direct comparison of AMBS with popular constrained RL algorithms. We also provide strong probabilistic safety guarantees for the continuous setting. In addition, we propose two novel penalty techniques that directly modify the policy gradient, which empirically provide more stable convergence in our experiments.

arXiv:2402.00355

Adaptive Primal-Dual Method for Safe Reinforcement Learning

Primal-dual methods have a natural application in Safe Reinforcement Learning (SRL), posed as a constrained policy optimization problem. In practice however, applying primal-dual methods to SRL is challenging, due to the inter-dependency of the learning rate (LR) and Lagrangian multipliers (dual variables) each time an embedded unconstrained RL problem is solved. In this paper, we propose, analyze and evaluate adaptive primal-dual (APD) methods for SRL, where two adaptive LRs are adjusted to the Lagrangian multipliers so as to optimize the policy in each iteration. We theoretically establish the convergence, optimality and feasibility of the APD algorithm. Finally, we conduct numerical evaluation of the practical APD algorithm with four well-known environments in Bullet-Safey-Gym employing two state-of-the-art SRL algorithms: PPO-Lagrangian and DDPG-Lagrangian. All experiments show that the practical APD algorithm outperforms (or achieves comparable performance) and attains more stable training than the constant LR cases. Additionally, we substantiate the robustness of selecting the two adaptive LRs by empirical evidence.

arXiv:2401.17837

Safe Reinforcement Learning-Based Eco-Driving Control for Mixed Traffic Flows With Disturbances

This paper presents a safe learning-based eco-driving framework tailored for mixed traffic flows, which aims to optimize energy efficiency while guaranteeing safety during real-system operations. Even though reinforcement learning (RL) is capable of optimizing energy efficiency in intricate environments, it is challenged by safety requirements during the training process. The lack of safety guarantees is the other concern when deploying a trained policy in real-world application. Compared with RL, model predicted control (MPC) can handle constrained dynamics systems, ensuring safe driving. However, the major challenges lie in complicated eco-driving tasks and the presence of disturbances, which respectively challenge the MPC design and the satisfaction of constraints. To address these limitations, the proposed framework incorporates the tube-based enhanced MPC (RMPC) to ensure the safe execution of the RL policy under disturbances, thereby improving the control robustness. RL not only optimizes the energy efficiency of the connected and automated vehicle in mixed traffic but also handles more uncertain scenarios, in which the energy consumption of the human-driven vehicle and its diverse and stochastic driving behaviors are considered in the optimization framework. Simulation results demonstrate that the proposed algorithm, compared with RMPC technique, shows an average improvement of 10.88% in holistic energy efficiency, while compared with RL algorithm, it effectively prevents inter-vehicle collisions.

arXiv:2401.14758공식 코드 · Papers with Code

Off-Policy Primal-Dual Safe Reinforcement Learning

Primal-dual safe RL methods commonly perform iterations between the primal update of the policy and the dual update of the Lagrange Multiplier. Such a training paradigm is highly susceptible to the error in cumulative cost estimation since this estimation serves as the key bond connecting the primal and dual update processes. We show that this problem causes significant underestimation of cost when using off-policy methods, leading to the failure to satisfy the safety constraint. To address this issue, we propose conservative policy optimization, which learns a policy in a constraint-satisfying area by considering the uncertainty in cost estimation. This improves constraint satisfaction but also potentially hinders reward maximization. We then introduce local policy convexification to help eliminate such suboptimality by gradually reducing the estimation uncertainty. We provide theoretical interpretations of the joint coupling effect of these two ingredients and further verify them by extensive experiments. Results on benchmark tasks show that our method not only achieves an asymptotic performance comparable to state-of-the-art on-policy methods while using much fewer samples, but also significantly reduces constraint violation during training. Our code is available at https://github.com/ZifanWu/CAL.

arXiv:2401.13148공식 코드 · Papers with Code

Stable and Safe Human-aligned Reinforcement Learning through Neural Ordinary Differential Equations

Reinforcement learning (RL) excels in applications such as video games, but ensuring safety as well as the ability to achieve the specified goals remains challenging when using RL for real-world problems, such as human-aligned tasks where human safety is paramount. This paper provides safety and stability definitions for such human-aligned tasks, and then proposes an algorithm that leverages neural ordinary differential equations (NODEs) to predict human and robot movements and integrates the control barrier function (CBF) and control Lyapunov function (CLF) with the actor-critic method to help to maintain the safety and stability for human-aligned tasks. Simulation results show that the algorithm helps the controlled robot to reach the desired goal state with fewer safety violations and better sample efficiency compared to other methods in a human-aligned task.

arXiv:2401.13020

A Safe Reinforcement Learning Algorithm for Supervisory Control of Power Plants

Traditional control theory-based methods require tailored engineering for each system and constant fine-tuning. In power plant control, one often needs to obtain a precise representation of the system dynamics and carefully design the control scheme accordingly. Model-free Reinforcement learning (RL) has emerged as a promising solution for control tasks due to its ability to learn from trial-and-error interactions with the environment. It eliminates the need for explicitly modeling the environment's dynamics, which is potentially inaccurate. However, the direct imposition of state constraints in power plant control raises challenges for standard RL methods. To address this, we propose a chance-constrained RL algorithm based on Proximal Policy Optimization for supervisory control. Our method employs Lagrangian relaxation to convert the constrained optimization problem into an unconstrained objective, where trainable Lagrange multipliers enforce the state constraints. Our approach achieves the smallest distance of violation and violation rate in a load-follow maneuver for an advanced Nuclear Power Plant design.