We present an efficient reinforcement learning algorithm that learns the optimal admission control policy in a partially observable queueing network. Specifically, only the arrival and departure times from the network are observable, and optimality refers to the average holding/rejection cost in infinite horizon. While reinforcement learning in Partially Observable Markov Decision Processes (POMDP) is prohibitively expensive in general, we show that our algorithm has a regret that only depends sub-linearly on the maximal number of jobs in the network, $S$. In particular, in contrast with existing regret analyses, our regret bound does not depend on the diameter of the underlying Markov Decision Process (MDP), which in most queueing systems is at least exponential in $S$. The novelty of our approach is to leverage Norton's equivalent theorem for closed product-form queueing networks and an efficient reinforcement learning algorithm for MDPs with the structure of birth-and-death processes.

Applying Reinforcement Learning (RL) to sequence generation models enables the direct optimization of long-term rewards (\textit{e.g.,} BLEU and human feedback), but typically requires large-scale sampling over a space of action sequences. This is a computational challenge as presented by the practice of sequence generation problems, such as machine translation, where we often deal with a large action space (\textit{e.g.,} a vocabulary) and a long action sequence (\textit{e.g.,} a translation). In this work, we introduce two-stage sampling and dynamic sampling approaches to improve the sampling efficiency during training sequence generation models via RL. We experiment with our approaches on the traditional sequence generation tasks, including machine translation and abstractive summarization. Furthermore, we evaluate our approaches in RL from human feedback (RLHF) through training a large language model using the reward model. Experimental results show that the efficient sampling-based RL, referred to as ESRL, can outperform all baselines in terms of both training efficiency and memory consumption. Notably, ESRL yields consistent performance gains over the strong REINFORCE, minimum risk training, and proximal policy optimization methods.

This paper addresses the prevalent lack of tools to facilitate and empower Game Theory and Artificial Intelligence (AI) research in cybersecurity. The primary contribution is the introduction of ExploitFlow (EF), an AI and Game Theory-driven modular library designed for cyber security exploitation. EF aims to automate attacks, combining exploits from various sources, and capturing system states post-action to reason about them and understand potential attack trees. The motivation behind EF is to bolster Game Theory and AI research in cybersecurity, with robotics as the initial focus. Results indicate that EF is effective for exploring machine learning in robot cybersecurity. An artificial agent powered by EF, using Reinforcement Learning, outperformed both brute-force and human expert approaches, laying the path for using ExploitFlow for further research. Nonetheless, we identified several limitations in EF-driven agents, including a propensity to overfit, the scarcity and production cost of datasets for generalization, and challenges in interpreting networking states across varied security settings. To leverage the strengths of ExploitFlow while addressing identified shortcomings, we present Malism, our vision for a comprehensive automated penetration testing framework with ExploitFlow at its core.

Recent months have seen the emergence of a powerful new trend in which large language models (LLMs) are augmented to become autonomous language agents capable of performing objective oriented multi-step tasks on their own, rather than merely responding to queries from human users. Most existing language agents, however, are not optimized using environment-specific rewards. Although some agents enable iterative refinement through verbal feedback, they do not reason and plan in ways that are compatible with gradient-based learning from rewards. This paper introduces a principled framework for reinforcing large language agents by learning a retrospective model, which automatically tunes the language agent prompts from environment feedback through policy gradient. Specifically, our proposed agent architecture learns from rewards across multiple environments and tasks, for fine-tuning a pre-trained language model which refines the language agent prompt by summarizing the root cause of prior failed attempts and proposing action plans. Experimental results on various tasks demonstrate that the language agents improve over time and that our approach considerably outperforms baselines that do not properly leverage gradients from the environment. This demonstrates that using policy gradient optimization to improve language agents, for which we believe our work is one of the first, seems promising and can be applied to optimize other models in the agent architecture to enhance agent performances over time.

One major limitation to the applicability of Reinforcement Learning (RL) to many practical domains is the large number of samples required to learn an optimal policy. To address this problem and improve learning efficiency, we consider a linear hierarchy of abstraction layers of the Markov Decision Process (MDP) underlying the target domain. Each layer is an MDP representing a coarser model of the one immediately below in the hierarchy. In this work, we propose a novel form of Reward Shaping where the solution obtained at the abstract level is used to offer rewards to the more concrete MDP, in such a way that the abstract solution guides the learning in the more complex domain. In contrast with other works in Hierarchical RL, our technique has few requirements in the design of the abstract models and it is also tolerant to modeling errors, thus making the proposed approach practical. We formally analyze the relationship between the abstract models and the exploration heuristic induced in the lower-level domain. Moreover, we prove that the method guarantees optimal convergence and we demonstrate its effectiveness experimentally.

Deep Reinforcement Learning (DRL) algorithms have been increasingly employed during the last decade to solve various decision-making problems such as autonomous driving and robotics. However, these algorithms have faced great challenges when deployed in safety-critical environments since they often exhibit erroneous behaviors that can lead to potentially critical errors. One way to assess the safety of DRL agents is to test them to detect possible faults leading to critical failures during their execution. This raises the question of how we can efficiently test DRL policies to ensure their correctness and adherence to safety requirements. Most existing works on testing DRL agents use adversarial attacks that perturb states or actions of the agent. However, such attacks often lead to unrealistic states of the environment. Their main goal is to test the robustness of DRL agents rather than testing the compliance of agents' policies with respect to requirements. Due to the huge state space of DRL environments, the high cost of test execution, and the black-box nature of DRL algorithms, the exhaustive testing of DRL agents is impossible. In this paper, we propose a Search-based Testing Approach of Reinforcement Learning Agents (STARLA) to test the policy of a DRL agent by effectively searching for failing executions of the agent within a limited testing budget. We use machine learning models and a dedicated genetic algorithm to narrow the search towards faulty episodes. We apply STARLA on Deep-Q-Learning agents which are widely used as benchmarks and show that it significantly outperforms Random Testing by detecting more faults related to the agent's policy. We also investigate how to extract rules that characterize faulty episodes of the DRL agent using our search results. Such rules can be used to understand the conditions under which the agent fails and thus assess its deployment risks.

Deep reinforcement learning algorithms (DRL) are increasingly being used in safety-critical systems. Ensuring the safety of DRL agents is a critical concern in such contexts. However, relying solely on testing is not sufficient to ensure safety as it does not offer guarantees. Building safety monitors is one solution to alleviate this challenge. This paper proposes SMARLA, a machine learning-based safety monitoring approach designed for DRL agents. For practical reasons, SMARLA is designed to be black-box (as it does not require access to the internals of the agent) and leverages state abstraction to reduce the state space and thus facilitate the learning of safety violation prediction models from agent's states. We validated SMARLA on two well-known RL case studies. Empirical analysis reveals that SMARLA achieves accurate violation prediction with a low false positive rate, and can predict safety violations at an early stage, approximately halfway through the agent's execution before violations occur.

Text-adventure games and text role-playing games are grand challenges for reinforcement learning game playing agents. Text role-playing games are open-ended environments where an agent must faithfully play a particular character. We consider the distinction between characters and actors, where an actor agent has the ability to play multiple characters. We present a framework we call a thespian agent that can learn to emulate multiple characters along with a soft prompt that can be used to direct it as to which character to play at any time. We further describe an attention mechanism that allows the agent to learn new characters that are based on previously learned characters in a few-shot fashion. We show that our agent outperforms the state of the art agent framework in multi-character learning and few-shot learning.

Job scheduling is a well-known Combinatorial Optimization problem with endless applications. Well planned schedules bring many benefits in the context of automated systems: among others, they limit production costs and waste. Nevertheless, the NP-hardness of this problem makes it essential to use heuristics whose design is difficult, requires specialized knowledge and often produces methods tailored to the specific task. This paper presents an original end-to-end Deep Reinforcement Learning approach to scheduling that automatically learns dispatching rules. Our technique is inspired by natural language encoder-decoder models for sequence processing and has never been used, to the best of our knowledge, for scheduling purposes. We applied and tested our method in particular to some benchmark instances of Job Shop Problem, but this technique is general enough to be potentially used to tackle other different optimal job scheduling tasks with minimal intervention. Results demonstrate that we outperform many classical approaches exploiting priority dispatching rules and show competitive results on state-of-the-art Deep Reinforcement Learning ones.

Efficient exploration in complex environments remains Distributional RL (DiRL) has rapidly established its place a major challenge for reinforcement learning among reinforcement learning (RL) algorithms Bellemare (RL). Compared to previous Thompson samplinginspired et al. [2017] as a powerful improvement over nondistributional mechanisms that enable temporally extended value-based counterparts Lyle et al. [2019]. In exploration, i.e., deep exploration, we focus DiRL, the agent does not learn a single summary statistic of on deep exploration in distributional RL. We develop the return for each state-action pair, but instead learns the here a general purpose approach, Bag of Policies whole return distribution. The agent's behaviour is being (BoP), that can be built on top of any return evaluated for multiple possible consequences which in turn distribution estimator by maintaining a population affect the policy update. While this does lead to more stable of its copies. BoP consists of an ensemble of multiple learning and better performance Lyle et al. [2019], it does heads that are updated independently. During not itself change the way actions are selected; as distributional training, each episode is controlled by only one of extensions to value-based RL, in C51 Bellemare et al. the heads and the collected state-action pairs are [2017], QR-DQN Dabney et al. [2018b] the agent still takes used to update all heads off-policy, leading to distinct actions according to the mean of the estimated return distributions learning signals for each head which diversify in each state-action pair.