PyTorchKR Papers

최신 논문

총 319편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2110.15907

Learning to Be Cautious

A key challenge in the field of reinforcement learning is to develop agents that behave cautiously in novel situations. It is generally impossible to anticipate all situations that an autonomous system may face or what behavior would best avoid bad outcomes. An agent that can learn to be cautious would overcome this challenge by discovering for itself when and how to behave cautiously. In contrast, current approaches typically embed task-specific safety information or explicit cautious behaviors into the system, which is error-prone and imposes extra burdens on practitioners. In this paper, we present both a sequence of tasks where cautious behavior becomes increasingly non-obvious, as well as an algorithm to demonstrate that it is possible for a system to learn to be cautious. The essential features of our algorithm are that it characterizes reward function uncertainty without task-specific safety information and uses this uncertainty to construct a robust policy. Specifically, we construct robust policies with a k-of-N counterfactual regret minimization (CFR) subroutine given learned reward function uncertainty represented by a neural network ensemble. These policies exhibit caution in each of our tasks without any task-specific safety tuning.

arXiv:2110.14468

DESTA: A Framework for Safe Reinforcement Learning with Markov Games of Intervention

Reinforcement learning (RL) involves performing exploratory actions in an unknown system. This can place a learning agent in dangerous and potentially catastrophic system states. Current approaches for tackling safe learning in RL simultaneously trade-off safe exploration and task fulfillment. In this paper, we introduce a new generation of RL solvers that learn to minimise safety violations while maximising the task reward to the extent that can be tolerated by the safe policy. Our approach introduces a novel two-player framework for safe RL called Distributive Exploration Safety Training Algorithm (DESTA). The core of DESTA is a game between two adaptive agents: Safety Agent that is delegated the task of minimising safety violations and Task Agent whose goal is to maximise the environment reward. Specifically, Safety Agent can selectively take control of the system at any given point to prevent safety violations while Task Agent is free to execute its policy at any other states. This framework enables Safety Agent to learn to take actions at certain states that minimise future safety violations, both during training and testing time, while Task Agent performs actions that maximise the task performance everywhere else. Theoretically, we prove that DESTA converges to stable points enabling safety violations of pretrained policies to be minimised. Empirically, we show DESTA's ability to augment the safety of existing policies and secondly, construct safe RL policies when the Task Agent and Safety Agent are trained concurrently. We demonstrate DESTA's superior performance against leading RL methods in Lunar Lander and Frozen Lake from OpenAI gym.

arXiv:2110.10333

Computationally Efficient Safe Reinforcement Learning for Power Systems

We propose a computationally efficient approach to safe reinforcement learning (RL) for frequency regulation in power systems with high levels of variable renewable energy resources. The approach draws on set-theoretic control techniques to craft a neural network-based control policy that is guaranteed to satisfy safety-critical state constraints, without needing to solve a model predictive control or projection problem in real time. By exploiting the properties of robust controlled-invariant polytopes, we construct a novel, closed-form "safety-filter" that enables end-to-end safe learning using any policy gradient-based RL algorithm. We then apply the safety filter in conjunction with the deep deterministic policy gradient (DDPG) algorithm to regulate frequency in a modified 9-bus power system, and show that the learned policy is more cost-effective than robust linear feedback control techniques while maintaining the same safety guarantee. We also show that the proposed paradigm outperforms DDPG augmented with constraint violation penalties.

arXiv:2110.07699

Safe Autonomous Racing via Approximate Reachability on Ego-vision

Racing demands each vehicle to drive at its physical limits, when any safety infraction could lead to catastrophic failure. In this work, we study the problem of safe reinforcement learning (RL) for autonomous racing, using the vehicle's ego-camera view and speed as input. Given the nature of the task, autonomous agents need to be able to 1) identify and avoid unsafe scenarios under the complex vehicle dynamics, and 2) make sub-second decision in a fast-changing environment. To satisfy these criteria, we propose to incorporate Hamilton-Jacobi (HJ) reachability theory, a safety verification method for general non-linear systems, into the constrained Markov decision process (CMDP) framework. HJ reachability not only provides a control-theoretic approach to learn about safety, but also enables low-latency safety verification. Though HJ reachability is traditionally not scalable to high-dimensional systems, we demonstrate that with neural approximation, the HJ safety value can be learned directly on vision context -- the highest-dimensional problem studied via the method, to-date. We evaluate our method on several benchmark tasks, including Safety Gym and Learn-to-Race (L2R), a recently-released high-fidelity autonomous racing environment. Our approach has significantly fewer constraint violations in comparison to other constrained RL baselines in Safety Gym, and achieves the new state-of-the-art results on the L2R benchmark task. We provide additional visualization of agent behavior at the following anonymized paper website: https://sites.google.com/view/safeautonomousracing/home

arXiv:2110.05415공식 코드 · Papers with Code

Safe Reinforcement Learning Using Robust Control Barrier Functions

Reinforcement Learning (RL) has been shown to be effective in many scenarios. However, it typically requires the exploration of a sufficiently large number of state-action pairs, some of which may be unsafe. Consequently, its application to safety-critical systems remains a challenge. An increasingly common approach to address safety involves the addition of a safety layer that projects the RL actions onto a safe set of actions. In turn, a difficulty for such frameworks is how to effectively couple RL with the safety layer to improve the learning performance. In this paper, we frame safety as a differentiable robust-control-barrier-function layer in a model-based RL framework. Moreover, we also propose an approach to modularly learn the underlying reward-driven task, independent of safety constraints. We demonstrate that this approach both ensures safety and effectively guides exploration during training in a range of experiments, including zero-shot transfer when the reward is learned in a modular way.

arXiv:2110.01126

Decentralized Safe Reinforcement Learning for Voltage Control

Inverter-based distributed energy resources provide the possibility for fast time-scale voltage control by quickly adjusting their reactive power. The power-electronic interfaces allow these resources to realize almost arbitrary control law, but designing these decentralized controllers is nontrivial. Reinforcement learning (RL) approaches are becoming increasingly popular to search for policy parameterized by neural networks. It is difficult, however, to enforce that the learned controllers are safe, in the sense that they may introduce instabilities into the system. This paper proposes a safe learning approach for voltage control. We prove that the system is guaranteed to be exponentially stable if each controller satisfies certain Lipschitz constraints. The set of Lipschitz bound is optimized to enlarge the search space for neural network controllers. We explicitly engineer the structure of neural network controllers such that they satisfy the Lipschitz constraints by design. A decentralized RL framework is constructed to train local neural network controller at each bus in a model-free setting.

arXiv:2109.12674공식 코드 · Papers with Code

MetaDrive: Composing Diverse Driving Scenarios for Generalizable Reinforcement Learning

Driving safely requires multiple capabilities from human and intelligent agents, such as the generalizability to unseen environments, the safety awareness of the surrounding traffic, and the decision-making in complex multi-agent settings. Despite the great success of Reinforcement Learning (RL), most of the RL research works investigate each capability separately due to the lack of integrated environments. In this work, we develop a new driving simulation platform called MetaDrive to support the research of generalizable reinforcement learning algorithms for machine autonomy. MetaDrive is highly compositional, which can generate an infinite number of diverse driving scenarios from both the procedural generation and the real data importing. Based on MetaDrive, we construct a variety of RL tasks and baselines in both single-agent and multi-agent settings, including benchmarking generalizability across unseen scenes, safe exploration, and learning multi-agent traffic. The generalization experiments conducted on both procedurally generated scenarios and real-world scenarios show that increasing the diversity and the size of the training set leads to the improvement of the RL agent's generalizability. We further evaluate various safe reinforcement learning and multi-agent reinforcement learning algorithms in MetaDrive environments and provide the benchmarks. Source code, documentation, and demo video are available at \url{ https://metadriverse.github.io/metadrive}.

arXiv:2109.05077

Data Generation Method for Learning a Low-dimensional Safe Region in Safe Reinforcement Learning

Safe reinforcement learning aims to learn a control policy while ensuring that neither the system nor the environment gets damaged during the learning process. For implementing safe reinforcement learning on highly nonlinear and high-dimensional dynamical systems, one possible approach is to find a low-dimensional safe region via data-driven feature extraction methods, which provides safety estimates to the learning algorithm. As the reliability of the learned safety estimates is data-dependent, we investigate in this work how different training data will affect the safe reinforcement learning approach. By balancing between the learning performance and the risk of being unsafe, a data generation method that combines two sampling methods is proposed to generate representative training data. The performance of the method is demonstrated with a three-link inverted pendulum example.

arXiv:2108.06266공식 코드 · Papers with Code

Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learning

The last half-decade has seen a steep rise in the number of contributions on safe learning methods for real-world robotic deployments from both the control and reinforcement learning communities. This article provides a concise but holistic review of the recent advances made in using machine learning to achieve safe decision making under uncertainties, with a focus on unifying the language and frameworks used in control theory and reinforcement learning research. Our review includes: learning-based control approaches that safely improve performance by learning the uncertain dynamics, reinforcement learning approaches that encourage safety or robustness, and methods that can formally certify the safety of a learned control policy. As data- and learning-based robot control methods continue to gain traction, researchers must understand when and how to best leverage them in real-world scenarios where safety is imperative, such as when operating in close proximity to humans. We highlight some of the open challenges that will drive the field of robot learning in the coming years, and emphasize the need for realistic physics-based benchmarks to facilitate fair comparisons between control and reinforcement learning approaches.

arXiv:2108.01846공식 코드 · Papers with Code

Learning Barrier Certificates: Towards Safe Reinforcement Learning with Zero Training-time Violations

Training-time safety violations have been a major concern when we deploy reinforcement learning algorithms in the real world. This paper explores the possibility of safe RL algorithms with zero training-time safety violations in the challenging setting where we are only given a safe but trivial-reward initial policy without any prior knowledge of the dynamics model and additional offline data. We propose an algorithm, Co-trained Barrier Certificate for Safe RL (CRABS), which iteratively learns barrier certificates, dynamics models, and policies. The barrier certificates, learned via adversarial training, ensure the policy's safety assuming calibrated learned dynamics model. We also add a regularization term to encourage larger certified regions to enable better exploration. Empirical simulations show that zero safety violations are already challenging for a suite of simple environments with only 2-4 dimensional state space, especially if high-reward policies have to visit regions near the safety boundary. Prior methods require hundreds of violations to achieve decent rewards on these tasks, whereas our proposed algorithms incur zero violations.

arXiv:2107.13944

Lyapunov-based uncertainty-aware safe reinforcement learning

Reinforcement learning (RL) has shown a promising performance in learning optimal policies for a variety of sequential decision-making tasks. However, in many real-world RL problems, besides optimizing the main objectives, the agent is expected to satisfy a certain level of safety (e.g., avoiding collisions in autonomous driving). While RL problems are commonly formalized as Markov decision processes (MDPs), safety constraints are incorporated via constrained Markov decision processes (CMDPs). Although recent advances in safe RL have enabled learning safe policies in CMDPs, these safety requirements should be satisfied during both training and in the deployment process. Furthermore, it is shown that in memory-based and partially observable environments, these methods fail to maintain safety over unseen out-of-distribution observations. To address these limitations, we propose a Lyapunov-based uncertainty-aware safe RL model. The introduced model adopts a Lyapunov function that converts trajectory-based constraints to a set of local linear constraints. Furthermore, to ensure the safety of the agent in highly uncertain environments, an uncertainty quantification method is developed that enables identifying risk-averse actions through estimating the probability of constraint violations. Moreover, a Transformers model is integrated to provide the agent with memory to process long time horizons of information via the self-attention mechanism. The proposed model is evaluated in grid-world navigation tasks where safety is defined as avoiding static and dynamic obstacles in fully and partially observable environments. The results of these experiments show a significant improvement in the performance of the agent both in achieving optimality and satisfying safety constraints.

arXiv:2107.12045공식 코드 · Papers with Code

How to Certify Machine Learning Based Safety-critical Systems? A Systematic Literature Review

Context: Machine Learning (ML) has been at the heart of many innovations over the past years. However, including it in so-called 'safety-critical' systems such as automotive or aeronautic has proven to be very challenging, since the shift in paradigm that ML brings completely changes traditional certification approaches. Objective: This paper aims to elucidate challenges related to the certification of ML-based safety-critical systems, as well as the solutions that are proposed in the literature to tackle them, answering the question 'How to Certify Machine Learning Based Safety-critical Systems?'. Method: We conduct a Systematic Literature Review (SLR) of research papers published between 2015 to 2020, covering topics related to the certification of ML systems. In total, we identified 217 papers covering topics considered to be the main pillars of ML certification: Robustness, Uncertainty, Explainability, Verification, Safe Reinforcement Learning, and Direct Certification. We analyzed the main trends and problems of each sub-field and provided summaries of the papers extracted. Results: The SLR results highlighted the enthusiasm of the community for this subject, as well as the lack of diversity in terms of datasets and type of models. It also emphasized the need to further develop connections between academia and industries to deepen the domain study. Finally, it also illustrated the necessity to build connections between the above mention main pillars that are for now mainly studied separately. Conclusion: We highlighted current efforts deployed to enable the certification of ML based software systems, and discuss some future research directions.

arXiv:2106.09110공식 코드 · Papers with Code

Safe Reinforcement Learning Using Advantage-Based Intervention

Many sequential decision problems involve finding a policy that maximizes total reward while obeying safety constraints. Although much recent research has focused on the development of safe reinforcement learning (RL) algorithms that produce a safe policy after training, ensuring safety during training as well remains an open problem. A fundamental challenge is performing exploration while still satisfying constraints in an unknown Markov decision process (MDP). In this work, we address this problem for the chance-constrained setting. We propose a new algorithm, SAILR, that uses an intervention mechanism based on advantage functions to keep the agent safe throughout training and optimizes the agent's policy using off-the-shelf RL algorithms designed for unconstrained MDPs. Our method comes with strong guarantees on safety during both training and deployment (i.e., after training and without the intervention mechanism) and policy performance compared to the optimal safety-constrained policy. In our experiments, we show that SAILR violates constraints far less during training than standard safe RL and constrained MDP approaches and converges to a well-performing policy that can be deployed safely without intervention. Our code is available at https://github.com/nolanwagener/safe_rl.

arXiv:2106.06239

Safe Reinforcement Learning with Linear Function Approximation

Safety in reinforcement learning has become increasingly important in recent years. Yet, existing solutions either fail to strictly avoid choosing unsafe actions, which may lead to catastrophic results in safety-critical systems, or fail to provide regret guarantees for settings where safety constraints need to be learned. In this paper, we address both problems by first modeling safety as an unknown linear cost function of states and actions, which must always fall below a certain threshold. We then present algorithms, termed SLUCB-QVI and RSLUCB-QVI, for episodic Markov decision processes (MDPs) with linear function approximation. We show that SLUCB-QVI and RSLUCB-QVI, while with \emph{no safety violation}, achieve a $\tilde{\mathcal{O}}\left(\kappa\sqrt{d^3H^3T}\right)$ regret, nearly matching that of state-of-the-art unsafe algorithms, where $H$ is the duration of each episode, $d$ is the dimension of the feature mapping, $\kappa$ is a constant characterizing the safety constraints, and $T$ is the total number of action plays. We further present numerical simulations that corroborate our theoretical findings.

arXiv:2105.10682공식 코드 · Papers with Code

Feasible Actor-Critic: Constrained Reinforcement Learning for Ensuring Statewise Safety

The safety constraints commonly used by existing safe reinforcement learning (RL) methods are defined only on expectation of initial states, but allow each certain state to be unsafe, which is unsatisfying for real-world safety-critical tasks. In this paper, we introduce the feasible actor-critic (FAC) algorithm, which is the first model-free constrained RL method that considers statewise safety, e.g, safety for each initial state. We claim that some states are inherently unsafe no matter what policy we choose, while for other states there exist policies ensuring safety, where we say such states and policies are feasible. By constructing a statewise Lagrange function available on RL sampling and adopting an additional neural network to approximate the statewise Lagrange multiplier, we manage to obtain the optimal feasible policy which ensures safety for each feasible state and the safest possible policy for infeasible states. Furthermore, the trained multiplier net can indicate whether a given state is feasible or not through the statewise complementary slackness condition. We provide theoretical guarantees that FAC outperforms previous expectation-based constrained RL methods in terms of both constraint satisfaction and reward optimization. Experimental results on both robot locomotive tasks and safe exploration tasks verify the safety enhancement and feasibility interpretation of the proposed method.

arXiv:2104.02547

Temporal-Logic-Based Intermittent, Optimal, and Safe Continuous-Time Learning for Trajectory Tracking

In this paper, we develop safe reinforcement-learning-based controllers for systems tasked with accomplishing complex missions that can be expressed as linear temporal logic specifications, similar to those required by search-and-rescue missions. We decompose the original mission into a sequence of tracking sub-problems under safety constraints. We impose the safety conditions by utilizing barrier functions to map the constrained optimal tracking problem in the physical space to an unconstrained one in the transformed space. Furthermore, we develop policies that intermittently update the control signal to solve the tracking sub-problems with reduced burden in the communication and computation resources. Subsequently, an actor-critic algorithm is utilized to solve the underlying Hamilton-Jacobi-Bellman equations. Finally, we support our proposed framework with stability proofs and showcase its efficacy via simulation results.

arXiv:2103.14186

Barrier Function-based Safe Reinforcement Learning for Emergency Control of Power Systems

Under voltage load shedding has been considered as a standard and effective measure to recover the voltage stability of the electric power grid under emergency and severe conditions. However, this scheme usually trips a massive amount of load which can be unnecessary and harmful to customers. Recently, deep reinforcement learning (RL) has been regarded and adopted as a promising approach that can significantly reduce the amount of load shedding. However, like most existing machine learning (ML)-based control techniques, RL control usually cannot guarantee the safety of the systems under control. In this paper, we introduce a novel safe RL method for emergency load shedding of power systems, that can enhance the safe voltage recovery of the electric power grid after experiencing faults. Unlike the standard RL method, the safe RL method has a reward function consisting of a Barrier function that goes to minus infinity when the system state goes to the safety bounds. Consequently, the optimal control policy, that maximizes the reward function, can render the power system to avoid the safety bounds. This method is general and can be applied to other safety-critical control problems. Numerical simulations on the 39-bus IEEE benchmark is performed to demonstrate the effectiveness of the proposed safe RL emergency control, as well as its adaptive capability to faults not seen in the training.

arXiv:2103.03656

Automatic Exploration Process Adjustment for Safe Reinforcement Learning with Joint Chance Constraint Satisfaction

In reinforcement learning (RL) algorithms, exploratory control inputs are used during learning to acquire knowledge for decision making and control, while the true dynamics of a controlled object is unknown. However, this exploring property sometimes causes undesired situations by violating constraints regarding the state of the controlled object. In this paper, we propose an automatic exploration process adjustment method for safe RL in continuous state and action spaces utilizing a linear nominal model of the controlled object. Specifically, our proposed method automatically selects whether the exploratory input is used or not at each time depending on the state and its predicted value as well as adjusts the variance-covariance matrix used in the Gaussian policy for exploration. We also show that our exploration process adjustment method theoretically guarantees the satisfaction of the constraints with the pre-specified probability, that is, the satisfaction of a joint chance constraint at every time. Finally, we illustrate the validity and the effectiveness of our method through numerical simulation.

arXiv:2103.01556공식 코드 · Papers with Code

Model-based Constrained Reinforcement Learning using Generalized Control Barrier Function

Model information can be used to predict future trajectories, so it has huge potential to avoid dangerous region when implementing reinforcement learning (RL) on real-world tasks, like autonomous driving. However, existing studies mostly use model-free constrained RL, which causes inevitable constraint violations. This paper proposes a model-based feasibility enhancement technique of constrained RL, which enhances the feasibility of policy using generalized control barrier function (GCBF) defined on the distance to constraint boundary. By using the model information, the policy can be optimized safely without violating actual safety constraints, and the sample efficiency is increased. The major difficulty of infeasibility in solving the constrained policy gradient is handled by an adaptive coefficient mechanism. We evaluate the proposed method in both simulations and real vehicle experiments in a complex autonomous driving collision avoidance task. The proposed method achieves up to four times fewer constraint violations and converges 3.36 times faster than baseline constrained RL approaches.

arXiv:2102.10643

Safe Reinforcement Learning Using Robust Action Governor

Reinforcement Learning (RL) is essentially a trial-and-error learning procedure which may cause unsafe behavior during the exploration-and-exploitation process. This hinders the application of RL to real-world control problems, especially to those for safety-critical systems. In this paper, we introduce a framework for safe RL that is based on integration of a RL algorithm with an add-on safety supervision module, called the Robust Action Governor (RAG), which exploits set-theoretic techniques and online optimization to manage safety-related requirements during learning. We illustrate this proposed safe RL framework through an application to automotive adaptive cruise control.