12 accepted papers
Many reinforcement learning algorithms, particularly those that rely on return estimates for policy improvement, can suffer from poor sample efficiency and training instability due to high-variance return estimates. In this paper we leverage new results from off-policy evaluation; it has recently be
Specifying informative and dense reward functions remains a pivotal challenge in Reinforcement Learning, as it directly affects the efficiency of agent training. In this work, we harness the expressive power of quantitative Linear Temporal Logic on finite traces to synthesize reward monitors that ge
In real-life scenarios, a Reinforcement Learning (RL) agent aiming to maximize their reward, must often also behave in a safe manner, including at training time. Thus, much attention in recent years has been given to Safe RL, where an agent aims to learn an optimal policy among all policies that sat…
We define maximum entropy goal-directedness (MEG), a formal measure of goal- directedness in causal models and Markov decision processes, and give algorithms for computing it. Measuring goal-directedness is important, as it is a critical element of many concerns about harm from AI. It is also of phi…
The behaviour of multi agent learning in competitive network games is often studied within the context of zero sum games, in which convergence guarantees may be obtained. However, outside of this class the behaviour of learning is known to display complex behaviours and convergence cannot be always…
We propose a new approach to the verification of epistemic properties of programmes. First, we introduce the new ``program-epistemic'' logic L_PK, which is strictly richer and more general than similar formalisms appearing in the literature. To solve the verification problem in an efficient way, we…
The behaviour of multi-agent learning in competitive settings is often considered under the restrictive assumption of a zero-sum game. Only under this strict requirement is the behaviour of learning well understood; beyond this, learning dynamics can often display non-convergent behaviours which pre…
Deceptive agents are a challenge for the safety, trustworthiness, and cooperation of AI systems. We focus on the problem that agents might deceive in order to achieve their goals (for instance, in our experiments with language models, the goal of being evaluated as truthful). There are a number of e…
The model checking problem for multi-agent systems against Strategy Logic specifications is known to be non-elementary. On this logic several fragments have been defined to tackle this issue but at the expense of expressiveness. In this paper, we propose a three-valued semantics for Strategy Logic u…
Understanding the impact of exploration on the behaviour of multi-agent learning has, so far, benefited from the restriction to potential, or network zero-sum games in which convergence to an equilibrium can be shown. Outside of these classes, learning dynamics rarely converge and little is known ab…
We address the problem of building agents whose goal is to learn to execute out-of distribution (OOD) multi-task instructions expressed in temporal logic (TL) by using deep reinforcement learning (DRL). Recent works provided evidence that the agent's neural architecture is a key feature when DRL age…
We study the semantics of knowledge in strategic reasoning. Most existing works either implicitly assume that agents do not know one another’s strategies, or that all strategies are known to all; and some works present inconsistent mixes of both features. We put forward a novel semantics for Strateg…