Beyond Expected Return: Accounting for Policy Reproducibility when Evaluating Reinforcement Learning Algorithms

Many applications in Reinforcement Learning (RL) usually have noise or stochasticity present in the environment. Beyond their impact on learning, these uncertainties lead the exact same policy to perform differently, i.e. yield different return, from one roll-out to another. Common evaluation procedures in RL summarise the consequent return distributions using solely the expected return, which does not account for the spread of the distribution. Our work defines this spread as the policy reproducibility: the ability of a policy to obtain similar performance when rolled out many times, a crucial property in some real-world applications. We highlight that existing procedures that only use the expected return are limited on two fronts: first an infinite number of return distributions with a wide range of performance-reproducibility trade-offs can have the same expected return, limiting its effectiveness when used for comparing policies; second, the expected return metric does not leave any room for practitioners to choose the best trade-off value for considered applications. In this work, we address these limitations by recommending the use of Lower Confidence Bound, a metric taken from Bayesian optimisation that provides the user with a preference parameter to choose a desired performance-reproducibility trade-off. We also formalise and quantify policy reproducibility, and demonstrate the benefit of our metrics using extensive experiments of popular RL algorithms on common uncertain RL tasks.

Mix-ME: Quality-Diversity for Multi-Agent Learning

In many real-world systems, such as adaptive robotics, achieving a single, optimised solution may be insufficient. Instead, a diverse set of high-performing solutions is often required to adapt to varying contexts and requirements. This is the realm of Quality-Diversity (QD), which aims to discover a collection of high-performing solutions, each with their own unique characteristics. QD methods have recently seen success in many domains, including robotics, where they have been used to discover damage-adaptive locomotion controllers. However, most existing work has focused on single-agent settings, despite many tasks of interest being multi-agent. To this end, we introduce Mix-ME, a novel multi-agent variant of the popular MAP-Elites algorithm that forms new solutions using a crossover-like operator by mixing together agents from different teams. We evaluate the proposed methods on a variety of partially observable continuous control tasks. Our evaluation shows that these multi-agent variants obtained by Mix-ME not only compete with single-agent baselines but also often outperform them in multi-agent settings under partial observability.

QDax: A Library for Quality-Diversity and Population-based Algorithms with Hardware Acceleration

QDax is an open-source library with a streamlined and modular API for Quality-Diversity (QD) optimization algorithms in Jax. The library serves as a versatile tool for optimization purposes, ranging from black-box optimization to continuous control. QDax offers implementations of popular QD, Neuroevolution, and Reinforcement Learning (RL) algorithms, supported by various examples. All the implementations can be just-in-time compiled with Jax, facilitating efficient execution across multiple accelerators, including GPUs and TPUs. These implementations effectively demonstrate the framework's flexibility and user-friendliness, easing experimentation for research purposes. Furthermore, the library is thoroughly documented and tested with 95\% coverage.

Benchmark tasks for Quality-Diversity applied to Uncertain domains

While standard approaches to optimisation focus on producing a single high-performing solution, Quality-Diversity (QD) algorithms allow large diverse collections of such solutions to be found. If QD has proven promising across a large variety of domains, it still struggles when faced with uncertain domains, where quantification of performance and diversity are non-deterministic. Previous work in Uncertain Quality-Diversity (UQD) has proposed methods and metrics designed for such uncertain domains. In this paper, we propose a first set of benchmark tasks to analyse and estimate the performance of UQD algorithms. We identify the key uncertainty properties to easily define UQD benchmark tasks: the uncertainty location, the type of distribution and its parameters. By varying the nature of those key UQD components, we introduce a set of 8 easy-to-implement and lightweight tasks, split into 3 main categories. All our tasks build on the Redundant Arm: a common QD environment that is lightweight and easily replicable. Each one of these tasks highlights one specific limitation that arises when considering UQD domains. With this first benchmark, we hope to facilitate later advances in UQD.

Don't Bet on Luck Alone: Enhancing Behavioral Reproducibility of Quality-Diversity Solutions in Uncertain Domains

Quality-Diversity (QD) algorithms are designed to generate collections of high-performing solutions while maximizing their diversity in a given descriptor space. However, in the presence of unpredictable noise, the fitness and descriptor of the same solution can differ significantly from one evaluation to another, leading to uncertainty in the estimation of such values. Given the elitist nature of QD algorithms, they commonly end up with many degenerate solutions in such noisy settings. In this work, we introduce Archive Reproducibility Improvement Algorithm (ARIA); a plug-and-play approach that improves the reproducibility of the solutions present in an archive. We propose it as a separate optimization module, relying on natural evolution strategies, that can be executed on top of any QD algorithm. Our module mutates solutions to (1) optimize their probability of belonging to their niche, and (2) maximize their fitness. The performance of our method is evaluated on various tasks, including a classical optimization problem and two high-dimensional control tasks in simulated robotic environments. We show that our algorithm enhances the quality and descriptor space coverage of any given archive by at least 50%.

Understanding the Synergies between Quality-Diversity and Deep Reinforcement Learning

The synergies between Quality-Diversity (QD) and Deep Reinforcement Learning (RL) have led to powerful hybrid QD-RL algorithms that have shown tremendous potential, and brings the best of both fields. However, only a single deep RL algorithm (TD3) has been used in prior hybrid methods despite notable progress made by other RL algorithms. Additionally, there are fundamental differences in the optimization procedures between QD and RL which would benefit from a more principled approach. We propose Generalized Actor-Critic QD-RL, a unified modular framework for actor-critic deep RL methods in the QD-RL setting. This framework provides a path to study insights from Deep RL in the QD-RL setting, which is an important and efficient way to make progress in QD-RL. We introduce two new algorithms, PGA-ME (SAC) and PGA-ME (DroQ) which apply recent advancements in Deep RL to the QD-RL setting, and solves the humanoid environment which was not possible using existing QD-RL algorithms. However, we also find that not all insights from Deep RL can be effectively translated to QD-RL. Critically, this work also demonstrates that the actor-critic models in QD-RL are generally insufficiently trained and performance gains can be achieved without any additional environment evaluations.

Multiple Hands Make Light Work: Enhancing Quality and Diversity using MAP-Elites with Multiple Parallel Evolution Strategies

With the development of hardware accelerators and their corresponding tools, evaluations have become more affordable through fast and massively parallel evaluations in some applications. This advancement has drastically sped up the runtime of evolution-inspired algorithms such as Quality-Diversity optimization, creating tremendous potential for algorithmic innovation through scale. In this work, we propose MAP-Elites-Multi-ES (MEMES), a novel QD algorithm based on Evolution Strategies (ES) designed for fast parallel evaluations. ME-Multi-ES builds on top of the existing MAP-Elites-ES algorithm, scaling it by maintaining multiple independent ES threads with massive parallelization. We also introduce a new dynamic reset procedure for the lifespan of the independent ES to autonomously maximize the improvement of the QD population. We show experimentally that MEMES outperforms existing gradient-based and objective-agnostic QD algorithms when compared in terms of generations. We perform this comparison on both black-box optimization and QD-Reinforcement Learning tasks, demonstrating the benefit of our approach across different problems and domains. Finally, we also find that our approach intrinsically enables optimization of fitness locally around a niche, a phenomenon not observed in other QD algorithms.

MAP-Elites with Descriptor-Conditioned Gradients and Archive Distillation into a Single Policy

Quality-Diversity algorithms, such as MAP-Elites, are a branch of Evolutionary Computation generating collections of diverse and high-performing solutions, that have been successfully applied to a variety of domains and particularly in evolutionary robotics. However, MAP-Elites performs a divergent search based on random mutations originating from Genetic Algorithms, and thus, is limited to evolving populations of low-dimensional solutions. PGA-MAP-Elites overcomes this limitation by integrating a gradient-based variation operator inspired by Deep Reinforcement Learning which enables the evolution of large neural networks. Although high-performing in many environments, PGA-MAP-Elites fails on several tasks where the convergent search of the gradient-based operator does not direct mutations towards archive-improving solutions. In this work, we present two contributions: (1) we enhance the Policy Gradient variation operator with a descriptor-conditioned critic that improves the archive across the entire descriptor space, (2) we exploit the actor-critic training to learn a descriptor-conditioned policy at no additional cost, distilling the knowledge of the archive into one single versatile policy that can execute the entire range of behaviors contained in the archive. Our algorithm, DCG-MAP-Elites improves the QD score over PGA-MAP-Elites by 82% on average, on a set of challenging locomotion tasks.

Uncertain Quality-Diversity: Evaluation methodology and new methods for Quality-Diversity in Uncertain Domains

Quality-Diversity optimisation (QD) has proven to yield promising results across a broad set of applications. However, QD approaches struggle in the presence of uncertainty in the environment, as it impacts their ability to quantify the true performance and novelty of solutions. This problem has been highlighted multiple times independently in previous literature. In this work, we propose to uniformise the view on this problem through four main contributions. First, we formalise a common framework for uncertain domains: the Uncertain QD setting, a special case of QD in which fitness and descriptors for each solution are no longer fixed values but distribution over possible values. Second, we propose a new methodology to evaluate Uncertain QD approaches, relying on a new per-generation sampling budget and a set of existing and new metrics specifically designed for Uncertain QD. Third, we propose three new Uncertain QD algorithms: Archive-sampling, Parallel-Adaptive-sampling and Deep-Grid-sampling. We propose these approaches taking into account recent advances in the QD community toward the use of hardware acceleration that enable large numbers of parallel evaluations and make sampling an affordable approach to uncertainty. Our final and fourth contribution is to use this new framework and the associated comparison methods to benchmark existing and novel approaches. We demonstrate once again the limitation of MAP-Elites in uncertain domains and highlight the performance of the existing Deep-Grid approach, and of our new algorithms. The goal of this framework and methods is to become an instrumental benchmark for future works considering Uncertain QD.

Efficient Exploration using Model-Based Quality-Diversity with Gradients

Exploration is a key challenge in Reinforcement Learning, especially in long-horizon, deceptive and sparse-reward environments. For such applications, population-based approaches have proven effective. Methods such as Quality-Diversity deals with this by encouraging novel solutions and producing a diversity of behaviours. However, these methods are driven by either undirected sampling (i.e. mutations) or use approximated gradients (i.e. Evolution Strategies) in the parameter space, which makes them highly sample-inefficient. In this paper, we propose a model-based Quality-Diversity approach. It extends existing QD methods to use gradients for efficient exploitation and leverage perturbations in imagination for efficient exploration. Our approach optimizes all members of a population simultaneously to maintain both performance and diversity efficiently by leveraging the effectiveness of QD algorithms as good data generators to train deep models. We demonstrate that it maintains the divergent search capabilities of population-based approaches on tasks with deceptive rewards while significantly improving their sample efficiency and quality of solutions.