Although parallelism has been extensively used in reinforcement learning (RL), the quantitative effects of parallel exploration are not well understood theoretically. We study the benefits of simple parallel exploration for reward-free RL in linear Markov decision processes (MDPs) and two-player zero-sum Markov games (MGs). In contrast to the existing literature, which focuses on approaches that encourage agents to explore a diverse set of policies, we show that using a single policy to guide exploration across all agents is sufficient to obtain an almost-linear speedup in all cases compared to their fully sequential counterpart. Furthermore, we demonstrate that this simple procedure is near-minimax optimal in the reward-free setting for linear MDPs. From a practical perspective, our paper shows that a single policy is sufficient and provably near-optimal for incorporating parallelism during the exploration phase.

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.

Learning control policies with large discrete action spaces is a challenging problem in the field of reinforcement learning due to present inefficiencies in exploration. With high dimensional action spaces, there are a large number of potential actions in each individual dimension over which policies would be learned. In this work, we introduce a Deep Reinforcement Learning (DRL) algorithm call Multi-Action Networks (MAN) Learning that addresses the challenge of high-dimensional large discrete action spaces. We propose factorizing the N-dimension action space into N 1-dimensional components, known as sub-actions, creating a Value Neural Network for each sub-action. Then, MAN uses temporal-difference learning to train the networks synchronously, which is simpler than training a single network with a large action output directly. To evaluate the proposed method, we test MAN on three scenarios: an n-dimension maze task, a block stacking task, and then extend MAN to handle 12 games from the Atari Arcade Learning environment with 18 action spaces. Our results indicate that MAN learns faster than both Deep Q-Learning and Double Deep Q-Learning, implying our method is a better performing synchronous temporal difference algorithm than those currently available for large discrete action spaces.

Deep reinforcement learning agents are notoriously sample inefficient, which considerably limits their application to real-world problems. Recently, many model-based methods have been designed to address this issue, with learning in the imagination of a world model being one of the most prominent approaches. However, while virtually unlimited interaction with a simulated environment sounds appealing, the world model has to be accurate over extended periods of time. Motivated by the success of Transformers in sequence modeling tasks, we introduce IRIS, a data-efficient agent that learns in a world model composed of a discrete autoencoder and an autoregressive Transformer. With the equivalent of only two hours of gameplay in the Atari 100k benchmark, IRIS achieves a mean human normalized score of 1.046, and outperforms humans on 10 out of 26 games, setting a new state of the art for methods without lookahead search. To foster future research on Transformers and world models for sample-efficient reinforcement learning, we release our code and models at https://github.com/eloialonso/iris.

In a special award session, the best papers of the conference were announced. The AAAI-2023 outstanding paper award went to Joar Skalse and Alessandro Abate for their work Misspecification in Inverse Reinforcement Learning. The AAAI-2023 outstanding student paper award was given to Decorate the Newcomers: Visual Domain Prompt for Continual Test Time Adaptation, authored by Yulu Gan, Yan Bai, Yihang Lou, Xianzheng Ma, Renrui Zhang, Nian Shi, and Lin Luo. There were also 12 distinguished paper award winners, the details of which can be found here. As well as these best paper awards, a number of prestigious AAAI awards were presented at the conference. These included the AAAI Award for Artificial Intelligence for the Benefit of Humanity, which was won by Tuomas Sandholm. You can find out more about this prize, and the others awarded, here. There will be plenty more content to come as we continue to cover the conference, and hear from participants about their work. You can find our conference coverage here, and this collection will be updated as soon as we add new content.

The rapid changes in the finance industry due to the increasing amount of data have revolutionized the techniques on data processing and data analysis and brought new theoretical and computational challenges. In contrast to classical stochastic control theory and other analytical approaches for solving financial decision-making problems that heavily reply on model assumptions, new developments from reinforcement learning (RL) are able to make full use of the large amount of financial data with fewer model assumptions and to improve decisions in complex financial environments. This survey paper aims to review the recent developments and use of RL approaches in finance. We give an introduction to Markov decision processes, which is the setting for many of the commonly used RL approaches. Various algorithms are then introduced with a focus on value and policy based methods that do not require any model assumptions. Connections are made with neural networks to extend the framework to encompass deep RL algorithms. Our survey concludes by discussing the application of these RL algorithms in a variety of decision-making problems in finance, including optimal execution, portfolio optimization, option pricing and hedging, market making, smart order routing, and robo-advising.

A common technique in reinforcement learning is to evaluate the value function from Monte Carlo simulations of a given policy, and use the estimated value function to obtain a new policy which is greedy with respect to the estimated value function. A well-known longstanding open problem in this context is to prove the convergence of such a scheme when the value function of a policy is estimated from data collected from a single sample path obtained from implementing the policy (see page 99 of [Sutton and Barto, 2018], page 8 of [Tsitsiklis, 2002]). We present a solution to the open problem by showing that a first-visit version of such a policy iteration scheme indeed converges to the optimal policy provided that the policy improvement step uses lookahead [Silver et al., 2016, Mnih et al., 2016, Silver et al., 2017b] rather than a simple greedy policy improvement. We provide results both for the original open problem in the tabular setting and also present extensions to the function approximation setting, where we show that the policy resulting from the algorithm performs close to the optimal policy within a function approximation error.

While deep reinforcement learning (RL) is becoming an integral part of good decision-making in data science, it is still plagued with sample inefficiency. This can be challenging when applying deep-RL in real-world environments where physical interactions are expensive and can risk system safety. To improve the sample efficiency of RL algorithms, this paper proposes a novel human-inspired framework that facilitates fast exploration and learning for difficult RL tasks. The main idea is to first provide the learning agent with simpler but similar tasks that gradually grow in difficulty and progress toward the main task. The proposed method requires no pre-training phase. Specifically, the learning of simpler tasks is only done for one iteration. The generated knowledge could be used by any transfer learning, including value transfer and policy transfer, to reduce the sample complexity while not adding to the computational complexity. So, it can be applied to any goal, environment, and reinforcement learning algorithm - both value-based methods and policy-based methods and both tabular methods and deep-RL methods. We have evaluated our proposed framework on both a simple Random Walk for illustration purposes and on more challenging optimal control problems with constraint. The experiments show the good performance of our proposed framework in improving the sample efficiency of RL-learning algorithms, especially when the main task is difficult.

This paper addresses the problem of visual feature representation learning with an aim to improve the performance of end-to-end reinforcement learning (RL) models. Specifically, a novel architecture is proposed that uses a heterogeneous loss function, called CRC loss, to learn improved visual features which can then be used for policy learning in RL. The CRC-loss function is a combination of three individual loss functions, namely, contrastive, reconstruction and consistency loss. The feature representation is learned in parallel to the policy learning while sharing the weight updates through a Siamese Twin encoder model. This encoder model is augmented with a decoder network and a feature projection network to facilitate computation of the above loss components. Through empirical analysis involving latent feature visualization, an attempt is made to provide an insight into the role played by this loss function in learning new action-dependent features and how they are linked to the complexity of the problems being solved. The proposed architecture, called CRC-RL, is shown to outperform the existing state-of-the-art methods on the challenging Deep mind control suite environments by a significant margin thereby creating a new benchmark in this field.

Nash Q-learning may be considered one of the first and most known algorithms in multi-agent reinforcement learning (MARL) for learning policies that constitute a Nash equilibrium of an underlying general-sum Markov game. Its original proof provided asymptotic guarantees and was for the tabular case. Recently, finite-sample guarantees have been provided using more modern RL techniques for the tabular case. Our work analyzes Nash Q-learning using linear function approximation -- a representation regime introduced when the state space is large or continuous -- and provides finite-sample guarantees that indicate its sample efficiency. We find that the obtained performance nearly matches an existing efficient result for single-agent RL under the same representation and has a polynomial gap when compared to the best-known result for the tabular case.