arXiv research

Revel tackles safe exploration in RL with verified symbolic policies.

SafePILCO is a Python tool for safe reinforcement learning.

Adapts safe policies for exploration in high-risk settings.

OSIL learns safe policies from unsafe demonstrations.

Previous work has shown the unreliability of existing algorithms in the batch Reinforcement Learning setting, and proposed the theoretically-grounded Safe Policy Improvement with Baseline Bootstrapping (SPIBB) fix: reproduce the baseline policy in the uncertain state-action pairs, in order to control the variance on th…

Improves policies with high certainty, even in small samples.

Safe reinforcement learning with logical constraints for optimal policy synthesis.

This paper addresses the question of how a previously available control policy $π_s$ can be used as a supervisor to more quickly and safely train a new learned control policy $π_L$ for a robot. A weighted average of the supervisor and learned policies is used during trials, with a heavier weight initially on the superv…

Algorithm safely learns from sub-optimal baseline policies while satisfying constraints.

We propose a method to optimise the parameters of a policy which will be used to safely perform a given task in a data-efficient manner. We train a Gaussian process model to capture the system dynamics, based on the PILCO framework. Our model has useful analytic properties, which allow closed form computation of error …

Safe-M$^3$-UCRL learns safe policies for multi-agent systems with global constraints.

Batch Reinforcement Learning (Batch RL) consists in training a policy using trajectories collected with another policy, called the behavioural policy. Safe policy improvement (SPI) provides guarantees with high probability that the trained policy performs better than the behavioural policy, also called baseline in this…

CoCoRL learns safe constraints from demonstrations with unknown rewards.

A new algorithm trains experts to safely guide agents in partially observed environments.

An important problem in sequential decision-making under uncertainty is to use limited data to compute a safe policy, i.e., a policy that is guaranteed to perform at least as well as a given baseline strategy. In this paper, we develop and analyze a new model-based approach to compute a safe policy when we have access …

A novel approach for safe offline RL using latent safety constraints.

Safe RL for autonomous vehicles using PCPO with trust regions and parallel learners.

Reinforcement learning is a powerful paradigm for learning optimal policies from experimental data. However, to find optimal policies, most reinforcement learning algorithms explore all possible actions, which may be harmful for real-world systems. As a consequence, learning algorithms are rarely applied on safety-crit…

New method tackles safe reinforcement learning from offline data.

Aims to create safe reinforcement learning policies by considering individual harm.

Safe autonomous decisions made with machine learning predictions using Conformal Decision Theory.

The paper provides a non-asymptotic error bound for linear system identification under nonlinear policies.

Algorithm finds safe zones in policy Markov Decision Processes to limit trajectory escape.

We study continuous action reinforcement learning problems in which it is crucial that the agent interacts with the environment only through safe policies, i.e.,~policies that do not take the agent to undesirable situations. We formulate these problems as constrained Markov decision processes (CMDPs) and present safe p…

Safe imitation learning with a safety layer for flexible training.

CRPO solves challenging SRL problems with convergence guarantee.

The paper tackles safe reinforcement learning with convex regularization.

New method provides scalable safety guarantees for RL agents.

We propose a framework for ensuring safe behavior of a reinforcement learning agent when the reward function may be difficult to specify. In order to do this, we rely on the existence of demonstrations from expert policies, and we provide a theoretical framework for the agent to optimize in the space of rewards consist…

We study the safe reinforcement learning problem with nonlinear function approximation, where policy optimization is formulated as a constrained optimization problem with both the objective and the constraint being nonconvex functions. For such a problem, we construct a sequence of surrogate convex constrained optimiza…

The paper tackles safe exploration in RL by a conservative safety critic.

SNPL learns safe policies for multi-objective interventions with high confidence.

New algorithm optimizes reward while ensuring safety in complex decision-making problems.

Safe exploration in RF-RL doesn't increase sample complexity.

In this work, we take a fresh look at some old and new algorithms for off-policy, return-based reinforcement learning. Expressing these in a common form, we derive a novel algorithm, Retrace($λ$), with three desired properties: (1) it has low variance; (2) it safely uses samples collected from any behaviour policy, wha…

SafeMIL learns safer policies by avoiding risky behavior from non-preferred trajectories.

Study assesses risks of European Safe Bonds using credit risk models.

Many currently deployed Reinforcement Learning agents work in an environment shared with humans, be them co-workers, users or clients. It is desirable that these agents adjust to people's preferences, learn faster thanks to their help, and act safely around them. We argue that most current approaches that learn from hu…

We show that when a third party, the adversary, steps into the two-party setting (agent and operator) of safely interruptible reinforcement learning, a trade-off has to be made between the probability of following the optimal policy in the limit, and the probability of escaping a dangerous situation created by the adve…

A new policy learning method allows policies to abstain when uncertain, improving safety and applicability.

ESRL uses uncertainty quantification to learn safe, optimal policies in offline RL.

Practical reinforcement learning problems are often formulated as constrained Markov decision process (CMDP) problems, in which the agent has to maximize the expected return while satisfying a set of prescribed safety constraints. In this study, we propose a novel simulator-based method to approximately solve a CMDP pr…

ASE safely explores unknown MDPs with unknown dynamics, improving sample efficiency.

New algorithm learns safe policies in unknown environments.

In this work we seek for an approach to integrate safety in the learning process that relies on a partly known state-space model of the system and regards the unknown dynamics as an additive bounded disturbance. We introduce a framework for safely learning a control strategy for a given system with an additive disturba…

This paper proposes a method to safely adjust exploration in RL to satisfy constraints.

We propose a policy improvement algorithm for Reinforcement Learning (RL) which is called Rerouted Behavior Improvement (RBI). RBI is designed to take into account the evaluation errors of the Q-function. Such errors are common in RL when learning the $Q$-value from finite past experience data. Greedy policies or even …

Proposes Constrained Q-learning for reinforcement learning with constraints.