Current reinforcement learning policies are heavily dependent upon the usage of a reward function. However, for advanced problems such as learning behavior for a self-driving car or a robot with many degrees of freedom, the reward function is very difficult or impossible to design to effectively encompass all human considerations. Inverse reinforcement learning algorithms are a class of algorithms that attempt to solve this issue by learning a reward function from expert demonstrations, and then subsequently using any manner of standard reinforcement learning algorithms to learn an optimal policy based off that learned reward function [1]. Inverse reinforcement learning algorithms require a dataset of expert demonstrations to derive a reward function from. However, this reward function can often be inaccurate and highly variable.

Policy-based algorithms are among the most widely adopted techniques in model-free RL, thanks to their strong theoretical groundings and good properties in continuous action spaces. Unfortunately, these methods require precise and problem-specific hyperparameter tuning to achieve good performance, and tend to struggle when asked to accomplish a series of heterogeneous tasks. In particular, the selection of the step size has a crucial impact on their ability to learn a highly performing policy, affecting the speed and the stability of the training process, and often being the main culprit for poor results. In this paper, we tackle these issues with a Meta Reinforcement Learning approach, by introducing a new formulation, known as meta-MDP, that can be used to solve any hyperparameter selection problem in RL with contextual processes. After providing a theoretical Lipschitz bound to the difference of performance in different tasks, we adopt the proposed framework to train a batch RL algorithm to dynamically recommend the most adequate step size for different policies and tasks. In conclusion, we present an experimental campaign to show the advantages of selecting an adaptive learning rate in heterogeneous environments.

Communication can impressively improve cooperation in multi-agent reinforcement learning (MARL), especially for partially-observed tasks. However, existing works either broadcast the messages leading to information redundancy, or learn targeted communication by modeling all the other agents as targets, which is not scalable when the number of agents varies. In this work, to tackle the scalability problem of MARL communication for partially-observed tasks, we propose a novel framework Transformer-based Email Mechanism (TEM). The agents adopt local communication to send messages only to the ones that can be observed without modeling all the agents. Inspired by human cooperation with email forwarding, we design message chains to forward information to cooperate with the agents outside the observation range. We introduce Transformer to encode and decode the message chain to choose the next receiver selectively. Empirically, TEM outperforms the baselines on multiple cooperative MARL benchmarks. When the number of agents varies, TEM maintains superior performance without further training.

Entropy regularized Markov decision processes have been widely used in reinforcement learning. This paper is concerned with the primal-dual formulation of the entropy regularized problems. Standard first-order methods suffer from slow convergence due to the lack of strict convexity and concavity. To address this issue, we first introduce a new quadratically convexified primal-dual formulation. The natural gradient ascent descent of the new formulation enjoys a global convergence guarantee and exponential convergence rate. We also propose a new interpolating metric that further accelerates the convergence significantly. Numerical results are provided to demonstrate the performance of the proposed methods under multiple settings.

Bandit and reinforcement learning (RL) algorithms demonstrated a wide range of successful real-life applications [Silver et al., 2016, 2017, Mnih et al., 2013, Berner et al., 2019, Vinyals et al., 2019, Mnih et al., 2015, Degrave et al., 2022]. Past works have theoretically studied the regret or sample complexity of various interactive decision making problems, such as contextual bandits, reinforcement learning (RL), partially observable Markov decision process (see Azar et al. [2017], Jin et al. [2018], Dong et al. [2021], Li et al. [2019], Agarwal et al. [2014], Foster and Rakhlin [2020], Jin et al. [2020], and references therein). Recently, Foster et al. [2021] present a unified algorithmic principle for achieving the minimax regret--the optimal regret for the worst-case problem instances. However, minimax regret bounds do not necessarily always present a full picture of the statistical complexity of the problem. They characterize the complexity of the most difficult instances, but potentially many other instances are much easier. An ideal algorithm should adapt to the complexity of a particular instance and incur smaller regrets on easy instances than the worst-case instances. Thus, an ideal regret bound should be instance-dependent, that is, depending on some properties of each instance. Prior works designed algorithms with instance-dependent regret bounds that are stronger than minimax regret bounds, but they are often not directly comparable because they depend on different properties of the instances, such as the gap conditions and the variance of the value function [Zanette and Brunskill, 2019, Xu et al., 2021, Foster et al., 2020, Tirinzoni et al., 2021]. A more ambitious and challenging goal is to design instance-optimal algorithms that outperform, on every instance, all consistent algorithms (those achieving non-trivial regrets on all instances).

Recent studies on visual reinforcement learning (visual RL) have explored the use of 3D visual representations. However, none of these work has systematically compared the efficacy of 3D representations with 2D representations across different tasks, nor have they analyzed 3D representations from the perspective of agent-object / object-object relationship reasoning. In this work, we seek answers to the question of when and how do 3D neural networks that learn features in the 3D-native space provide a beneficial inductive bias for visual RL. We specifically focus on 3D point clouds, one of the most common forms of 3D representations. We systematically investigate design choices for 3D point cloud RL, leading to the development of a robust algorithm for various robotic manipulation and control tasks. Furthermore, through comparisons between 2D image vs 3D point cloud RL methods on both minimalist synthetic tasks and complex robotic manipulation tasks, we find that 3D point cloud RL can significantly outperform the 2D counterpart when agent-object / object-object relationship encoding is a key factor.

We propose policy gradient algorithms for solving a risk-sensitive reinforcement learning (RL) problem in on-policy as well as off-policy settings. We consider episodic Markov decision processes, and model the risk using the broad class of smooth risk measures of the cumulative discounted reward. We propose two template policy gradient algorithms that optimize a smooth risk measure in on-policy and off-policy RL settings, respectively. We derive non-asymptotic bounds that quantify the rate of convergence of our proposed algorithms to a stationary point of the smooth risk measure. As special cases, we establish that our algorithms apply to optimization of mean-variance and distortion risk measures, respectively.

This work deals with the heterogeneous semantic-native communication (SNC) problem. When agents do not share the same communication context, the effectiveness of contextual reasoning (CR) is compromised calling for agents to infer other agents' context. This article proposes a novel framework for solving the inverse problem of CR in SNC using two Bayesian inference methods, namely: Bayesian inverse CR (iCR) and Bayesian inverse linearized CR (iLCR). The first proposed Bayesian iCR method utilizes Markov Chain Monte Carlo (MCMC) sampling to infer the agent's context while being computationally expensive. To address this issue, a Bayesian iLCR method is leveraged which obtains a linearized CR (LCR) model by training a linear neural network. Experimental results show that the Bayesian iLCR method requires less computation and achieves higher inference accuracy compared to Bayesian iCR. Additionally, heterogeneous SNC based on the context obtained through the Bayesian iLCR method shows better communication effectiveness than that of Bayesian iCR. Overall, this work provides valuable insights and methods to improve the effectiveness of SNC in situations where agents have different contexts.

Researchers have integrated exploration techniques into multi-agent reinforcement learning (MARL) algorithms, drawing on their remarkable success in deep reinforcement learning. Nonetheless, exploration in MARL presents a more substantial challenge, as agents need to coordinate their efforts in order to achieve comprehensive state coverage. Reaching a unanimous agreement on which kinds of states warrant exploring can be a struggle for agents in this context. We introduce \textbf{M}ulti-agent \textbf{E}xploration based on \textbf{S}ub-state \textbf{E}ntropy (MESE) to address this limitation. This novel approach incentivizes agents to explore states cooperatively by directing them to achieve consensus via an extra team reward. Calculating the additional reward is based on the novelty of the current sub-state that merits cooperative exploration. MESE employs a conditioned entropy approach to select the sub-state, using particle-based entropy estimation to calculate the entropy. MESE is a plug-and-play module that can be seamlessly integrated into most existing MARL algorithms, which makes it a highly effective tool for reinforcement learning. Our experiments demonstrate that MESE can substantially improve the MAPPO's performance on various tasks in the StarCraft multi-agent challenge (SMAC).

Capturing stochastic behaviors in business and work processes is essential to quantitatively understand how nondeterminism is resolved when taking decisions within the process. This is of special interest in process mining, where event data tracking the actual execution of the process are related to process models, and can then provide insights on frequencies and probabilities. Variants of stochastic Petri nets provide a natural formal basis for this. However, when capturing processes, such nets need to be labelled with (possibly duplicated) activities, and equipped with silent transitions that model internal, non-logged steps related to the orchestration of the process. At the same time, they have to be analyzed in a finite-trace semantics, matching the fact that each process execution consists of finitely many steps. These two aspects impede the direct application of existing techniques for stochastic Petri nets, calling for a novel characterization that incorporates labels and silent transitions in a finite-trace semantics. In this article, we provide such a characterization starting from generalized stochastic Petri nets and obtaining the framework of labelled stochastic processes (LSPs). On top of this framework, we introduce different key analysis tasks on the traces of LSPs and their probabilities. We show that all such analysis tasks can be solved analytically, in particular reducing them to a single method that combines automata-based techniques to single out the behaviors of interest within a LSP, with techniques based on absorbing Markov chains to reason on their probabilities. Finally, we demonstrate the significance of how our approach in the context of stochastic conformance checking, illustrating practical feasibility through a proof-of-concept implementation and its application to different datasets.