In this volume, we are happy present the post-proceedings of BNAIC/BeNeLearn 2023, the joint conference on Artificial Intelligence and Machine Learning in the BeNeLux, which took place at TU Delft. It is the main regional conference on these topics and has a long tradition: in 2018, the 30th Benelux Conference on Artificial Intelligence (BNAIC) and the 27th Belgian Dutch Conference on Machine Learning (Benelearn) were jointly organized in ‘s Hertogenbosch, and this has been repeated annually since. [...]@en

Model-based reinforcement learning (MBRL) has drawn considerable interest in recent years, given its promise to improve sample efficiency. Moreover, when using deep-learned models, it is possible to learn compact and generalizable models from data. In this work, we study MuZero, a state-of-the-art deep model-based reinforcement learning algorithm that distinguishes itself from existing algorithms by learning a value-equivalent model. Despite MuZero’s success and impact in the field of MBRL, existing literature has not thoroughly addressed why MuZero performs so well in practice. Specifically, there is a lack of in-depth investigation into the value-equivalent model learned by MuZero and its effectiveness in model-based credit assignment and policy improvement, which is vital for achieving sample efficiency in MBRL. To fill this gap, we explore two fundamental questions through our empirical analysis: 1) to what extent does MuZero achieve its learning objective of a value-equivalent model, and 2) how useful are these models for policy improvement? Among various other insights, we conclude that MuZero’s learned model cannot effectively generalize to evaluate unseen policies. This limitation constrains the extent to which we can additionally improve the current policy by planning with the model.@en

Game theory provides a mathematical way to study the interaction between multiple decision makers. However, classical game-theoretic analysis is limited in scalability due to the large number of strategies, precluding direct application to more complex scenarios. This survey provides a comprehensive overview of a framework for large games, known as Policy Space Response Oracles (PSRO), which holds promise to improve scalability by focusing attention on sufficient subsets of strategies. We first motivate PSRO and provide historical context. We then focus on the strategy exploration problem for PSRO: the challenge of assembling effective subsets of strategies that still represent the original game well with minimum computational cost. We survey current research directions for enhancing the efficiency of PSRO, and explore the applications of PSRO across various domains. We conclude by discussing open questions and future research. @en

Many methods for Model-based Reinforcement learning (MBRL) in Markov decision processes (MDPs) provide guarantees for both the accuracy of the model they can deliver and the learning efficiency. At the same time, state abstraction techniques allow for a reduction of the size of an MDP while maintaining a bounded loss with respect to the original problem. Therefore, it may come as a surprise that no such guarantees are available when combining both techniques, i.e., where MBRL merely observes abstract states. Our theoretical analysis shows that abstraction can introduce a dependence between samples collected online (e.g., in the real world). That means that, without taking this dependence into
account, results for MBRL do not directly extend to this setting. Our result shows that we can use concentration inequalities for martingales to overcome this problem. This result makes it possible to extend the guarantees of existing MBRL algorithms to the setting with abstraction. We illustrate this by combining R-MAX, a prototypical MBRL algorithm, with abstraction, thus producing the first performance guarantees for model-based ‘RL from Abstracted Observations’: model-based reinforcement learning with an abstract model.@en

Typically, a Reinforcement Learning (RL) algorithm focuses in learning a single deployable policy as the end product. Depending on the initialization methods and seed randomization, learning a single policy could possibly leads to convergence to different local optima across different runs, especially when the algorithm is sensitive to hyper-parameter tuning. Motivated by the capability of Generative Adversarial Networks (GANs) in learning complex data manifold, the adversarial training procedure could be utilized to learn a population of good-performing policies instead. We extend the teacher-student methodology observed in the Knowledge Distillation field in typical deep neural network prediction tasks to RL paradigm. Instead of learning a single compressed student network, an adversarially-trained generative model (hypernetwork) is learned to output network weights of a population of good-performing policy networks, representing a school of apprentices. Our proposed framework, named Teacher-Apprentices RL (TARL), is modular and could be used in conjunction with many existing RL algorithms. We illustrate the performance gain and improved robustness by combining TARL with various types of RL algorithms, including direct policy search Cross-Entropy Method, Q-learning, Actor-Critic, and policy gradient-based methods.

@en

Reinforcement learning agents may sometimes develop habits that are effective
only when specific policies are followed. After an initial exploration phase in which
agents try out different actions, they eventually converge toward a particular policy.
When this occurs, the distribution of state-action trajectories becomes narrower,
and agents start experiencing the same transitions again and again. At this point,
spurious correlations may arise. Agents may then pick up on these correlations
and learn state representations that do not generalize beyond the agent’s trajectory
distribution. In this paper, we provide a mathematical characterization of this
phenomenon, which we refer to as policy confounding, and show, through a series
of examples, when and how it occurs in practice.@en

One of the main challenges of multi-agent learning lies in establishing convergence of the algorithms, as, in general, a collection of individual, self-serving agents is not guaranteed to converge with their joint policy, when learning concurrently. This is in stark contrast to most single-agent environments, and sets a prohibitive barrier for deployment in practical applications, as it induces uncertainty in long term behavior of the system. In this work, we apply the concept of trapping regions, known from qualitative theory of dynamical systems, to create safety sets in the joint strategy space for decentralized learning. We propose a binary partitioning algorithm for verification that candidate sets form trapping regions in systems with known learning dynamics, and a heuristic sampling algorithm for scenarios where learning dynamics are not known. We demonstrate the applications to a regularized version of Dirac Generative Adversarial Network, a four-intersection traffic control scenario run in a state of the art open-source microscopic traffic simulator SUMO, and a mathematical model of economic competition.

@en

We present a review that unifies decision-support methods for exploring the solutions produced by multi-objective optimization (MOO) algorithms. As MOO is applied to solve diverse problems, approaches for analyzing the trade-offs offered by MOO algorithms are scattered across fields. We provide an overview of the advances on this topic, including methods for visualization, mining the solution set, and uncertainty exploration as well as emerging research directions, including interactivity, explainability, and ethics. We synthesize these methods drawing from different fields of research to build a unified approach, independent of the application. Our goals are to reduce the entry barrier for researchers and practitioners on using MOO algorithms and to provide novel research directions.@en

One of the main challenges of multi-agent learning lies in establishing convergence of the algorithms, as, in general, a collection of individual, self-serving agents is not guaranteed to converge with their joint policy, when learning concurrently. This is in stark contrast to most single-agent environments, and sets a prohibitive barrier for deployment in practical applications, as it induces uncertainty in long term behavior of the system. In this work, we propose to apply the concept of trapping regions, known from qualitative theory of dynamical systems, to create safety sets in the joint strategy space for decentralized learning. Upon verification of the direction of learning dynamics, the resulting trajectories are guaranteed not to escape such sets, during the learning process. As a result, it is ensured, that despite the uncertainty over convergence of the applied algorithms, learning will never form hazardous joint strategy combinations.

@en

This work investigates formal generalization error bounds that apply to support vector machines (SVMs) in realizable and agnostic learning problems. We focus on recently observed parallels between probably approximately correct (PAC)-learning bounds, such as compression and complexity-based bounds, and novel error guarantees derived within scenario theory. Scenario theory provides nonasymptotic and distributional-free error bounds for models trained by solving data-driven decision-making problems. Relevant theorems and assumptions are reviewed and discussed. We propose a numerical comparison of the tightness and effectiveness of theoretical error bounds for support vector classifiers trained on several randomized experiments from 13 real-life problems. This analysis allows for a fair comparison of different approaches from both conceptual and experimental standpoints. Based on the numerical results, we argue that the error guarantees derived from scenario theory are often tighter for realizable problems and always yield informative results, i.e., probability bounds tighter than a vacuous [0, 1] interval. This work promotes scenario theory as an alternative tool for model selection, structural-risk minimization, and generalization error analysis of SVMs. In this way, we hope to bring the communities of scenario and statistical learning theory closer, so that they can benefit from each other's insights.

@en

Model-based reinforcement learning methods are promising since they can increase sample efficiency while simultaneously improving generalizability. Learning can also be made more efficient through state abstraction, which delivers more compact models. Model-based reinforcement learning methods have been combined with learning abstract models to profit from both effects. We consider a wide range of state abstractions that have been covered in the literature, from straightforward state aggregation to deep learned representations, and sketch challenges that arise when combining model-based reinforcement learning with abstraction. We further show how various methods deal with these challenges and point to open questions and opportunities for further research.@en

Inferring reward functions from demonstrations and pairwise preferences are auspicious approaches for aligning Reinforcement Learning (RL) agents with human intentions. However, state-of-the art methods typically focus on learning a single reward model, thus rendering it difficult to trade off different reward functions from multiple experts. We propose Multi-Objective Reinforced Active Learning (MORAL), a novel method for combining diverse demonstrations of social norms into a Pareto-optimal policy. Through maintaining a distribution over scalarization weights, our approach is able to interactively tune a deep RL agent towards a variety of preferences, while eliminating the need for computing multiple policies. We empirically demonstrate the effectiveness of MORAL in two scenarios, which model a delivery and an emergency task that require an agent to act in the presence of normative conflicts. Overall, we consider our research a step towards multi-objective RL with learned rewards, bridging the gap between current reward learning and machine ethics literature.@en

Learning effective policies for real-world problems is still an open challenge for the field of reinforcement learning (RL). The main limitation being the amount of data needed and the pace at which that data can be obtained. In this paper, we study how to build lightweight simulators of complicated systems that can run sufficiently fast for deep RL to be applicable. We focus on domains where agents interact with a reduced portion of a larger environment while still being affected by the global dynamics. Our method combines the use of local simulators with learned models that mimic the influence of the global system. The experiments reveal that incorporating this idea into the deep RL workflow can considerably accelerate the training process and presents several opportunities for the future.

@en

This paper introduces Multi-Agent MDP Homomorphic Networks, a class of networks that allows distributed execution using only local information, yet is able to share experience between global symmetries in the joint state-action space of cooperative multi-agent systems. In cooperative multi-agent systems, complex symmetries arise between different configurations of the agents and their local observations. For example, consider a group of agents navigating: rotating the state globally results in a permutation of the optimal joint policy. Existing work on symmetries in single agent reinforcement learning can only be generalized to the fully centralized setting, because such approaches rely on the global symmetry in the full state-action spaces, and these can result in correspondences across agents. To encode such symmetries while still allowing distributed execution we propose a factorization that decomposes global symmetries into local transformations. Our proposed factorization allows for distributing the computation that enforces global symmetries over local agents and local interactions. We introduce a multi-agent equivariant policy network based on this factorization. We show empirically on symmetric multi-agent problems that globally symmetric distributable policies improve data efficiency compared to non-equivariant baselines.

@en

Learning effective policies for real-world problems is still an open challenge for the field of reinforcement learning (RL). The main limitation being the amount of data needed and the pace at which that data can be obtained. In this paper, we study how to build lightweight simulators of complicated systems that can run sufficiently fast for deep RL to be applicable. We focus on domains where agents interact with a reduced portion of a larger environment while still being affected by the global dynamics. Our method combines the use of local simulators with learned models that mimic the influence of the global system. The experiments reveal that incorporating this idea into the deep RL workflow can considerably accelerate the training process and presents several opportunities for the future.@en

While reinforcement learning (RL) has made great advances in scalability, exploration and partial observability are still active research topics. In contrast, Bayesian RL (BRL) provides a principled answer to both state estimation and the exploration-exploitation trade-off, but struggles to scale. To tackle this challenge, BRL frameworks with various prior assumptions have been proposed, with varied success. This work presents a representation-agnostic formulation of BRL under partially observability, unifying the previous models under one theoretical umbrella. To demonstrate its practical significance we also propose a novel derivation, Bayes-Adaptive Deep Dropout rl (BADDr), based on dropout networks. Under this parameterization, in contrast to previous work, the belief over the state and dynamics is a more scalable inference problem. We choose actions through Monte-Carlo tree search and empirically show that our method is competitive with state-of-the-art BRL methods on small domains while being able to solve much larger ones.

@en

Centaurs are half-human, half-AI decision-makers where the AI's goal is to complement the human. To do so, the AI must be able to recognize the goals and constraints of the human and have the means to help them. We present a novel formulation of the interaction between the human and the AI as a sequential game where the agents are modelled using Bayesian best-response models. We show that in this case the AI's problem of helping bounded-rational humans make better decisions reduces to a Bayes-adaptive POMDP. In our simulated experiments, we consider an instantiation of our framework for humans who are subjectively optimistic about the AI's future behaviour. Our results show that when equipped with a model of the human, the AI can infer the human's bounds and nudge them towards better decisions. We discuss ways in which the machine can learn to improve upon its own limitations as well with the help of the human. We identify a novel trade-off for centaurs in partially observable tasks: for the AI's actions to be acceptable to the human, the machine must make sure their beliefs are sufficiently aligned, but aligning beliefs might be costly. We present a preliminary theoretical analysis of this trade-off and its dependence on task structure.

@en

Learning to cooperate with other agents is challenging when those agents also possess the ability to adapt to our own behavior. Practical and theoretical approaches to learning in cooperative settings typically assume that other agents' behaviors are stationary, or else make very specific assumptions about other agents' learning processes. The goal of this work is to understand whether we can reliably learn to cooperate with other agents without such restrictive assumptions, which are unlikely to hold in real-world applications. Our main contribution is a set of impossibility results, which show that no learning algorithm can reliably learn to cooperate with all possible adaptive partners in a repeated matrix game, even if that partner is guaranteed to cooperate with some stationary strategy. Motivated by these results, we then discuss potential alternative assumptions which capture the idea that an adaptive partner will only adapt rationally to our behavior.

@en

Policy gradient methods have become one of the most popular classes of algorithms for multi-agent reinforcement learning. A key challenge, however, that is not addressed by many of these methods is multi-agent credit assignment: assessing an agent’s contribution to the overall performance, which is crucial for learning good policies. We propose a novel algorithm called Dr.Reinforce that explicitly tackles this by combining difference rewards with policy gradients to allow for learning decentralized policies when the reward function is known. By differencing the reward function directly, Dr.Reinforce avoids difficulties associated with learning the Q-function as done by counterfactual multi-agent policy gradients (COMA), a state-of-the-art difference rewards method. For applications where the reward function is unknown, we show the effectiveness of a version of Dr.Reinforce that learns an additional reward network that is used to estimate the difference rewards.

@en