We propose a new Q-learning variant, called 2RA Q-learning, that addresses some weaknesses of existing Q-learning methods in a principled manner. One such weakness is an underlying estimation bias which cannot be controlled and often results in poor performance. We propose a distributionally robust estimator for the maximum expected value term, which allows us to precisely control the level of estimation bias introduced. The distributionally robust estimator admits a closed-form solution such that the proposed algorithm has a computational cost per iteration comparable to Watkins' Q-learning. For the tabular case, we show that 2RA Q-learning converges to the optimal policy and analyze its asymptotic mean-squared error. Lastly, we conduct numerical experiments for various settings, which corroborate our theoretical findings and indicate that 2RA Q-learning often performs better than existing methods.

With economic development, the complexity of infrastructure has increased drastically. Similarly, with the shift from fossil fuels to renewable sources of energy, there is a dire need for such systems that not only predict and forecast with accuracy but also help in understanding the process of predictions. Artificial intelligence and machine learning techniques have helped in finding out wellperforming solutions to different problems in the energy sector. However, the usage of state-of-the-art techniques like reinforcement learning is not surprisingly convincing. This paper discusses the application of reinforcement techniques in energy systems and how explanations of these models can be helpful

Designing Reinforcement Learning (RL) solutions for real-life problems remains a significant challenge. A major area of concern is safety. "Shielding" is a popular technique to enforce safety in RL by turning user-defined safety specifications into safe agent behavior. However, these methods either suffer from extreme learning delays, demand extensive human effort in designing models and safe domains in the problem, or require pre-computation. In this paper, we propose a new permissibility-based framework to deal with safety and shield construction. Permissibility was originally designed for eliminating (non-permissible) actions that will not lead to an optimal solution to improve RL training efficiency. This paper shows that safety can be naturally incorporated into this framework, i.e. extending permissibility to include safety, and thereby we can achieve both safety and improved efficiency. Experimental evaluation using three standard RL applications shows the effectiveness of the approach.

Continual learning with deep neural networks presents challenges distinct from both the fixed-dataset and convex continual learning regimes. One such challenge is plasticity loss, wherein a neural network trained in an online fashion displays a degraded ability to fit new tasks. This problem has been extensively studied in both supervised learning and off-policy reinforcement learning (RL), where a number of remedies have been proposed. Still, plasticity loss has received less attention in the on-policy deep RL setting. Here we perform an extensive set of experiments examining plasticity loss and a variety of mitigation methods in on-policy deep RL. We demonstrate that plasticity loss is pervasive under domain shift in this regime, and that a number of methods developed to resolve it in other settings fail, sometimes even resulting in performance that is worse than performing no intervention at all. In contrast, we find that a class of ``regenerative'' methods are able to consistently mitigate plasticity loss in a variety of contexts, including in gridworld tasks and more challenging environments like Montezuma's Revenge and ProcGen.

The dominant framework for alignment of large language models (LLM), whether through reinforcement learning from human feedback or direct preference optimisation, is to learn from preference data. This involves building datasets where each element is a quadruplet composed of a prompt, two independent responses (completions of the prompt) and a human preference between the two independent responses, yielding a preferred and a dis-preferred response. Such data is typically scarce and expensive to collect. On the other hand, \emph{single-trajectory} datasets where each element is a triplet composed of a prompt, a response and a human feedback is naturally more abundant. The canonical element of such datasets is for instance an LLM's response to a user's prompt followed by a user's feedback such as a thumbs-up/down. Consequently, in this work, we propose DRO, or \emph{Direct Reward Optimisation}, as a framework and associated algorithms that do not require pairwise preferences. DRO uses a simple mean-squared objective that can be implemented in various ways. We validate our findings empirically, using T5 encoder-decoder language models, and show DRO's performance over selected baselines such as Kahneman-Tversky Optimization (KTO). Thus, we confirm that DRO is a simple and empirically compelling method for single-trajectory policy optimisation.

In reinforcement learning, it is often difficult to automate high-dimensional, rapid decision-making in dynamic environments, especially when domains require real-time online interaction and adaptive strategies such as web-based games. This work proposes a state-of-the-art Flexible Deep Q-Network (FDQN) framework that can address this challenge with a selfadaptive approach that is processing high-dimensional sensory data in realtime using a CNN and dynamically adapting the model architecture to varying action spaces of different gaming environments and outperforming previous baseline models in various Atari games and the Chrome Dino game as baselines. Using the epsilon-greedy policy, it effectively balances the new learning and exploitation for improved performance, and it has been designed with a modular structure that it can be easily adapted to other HTML-based games without touching the core part of the framework. It is demonstrated that the FDQN framework can successfully solve a well-defined task in a laboratory condition, but more importantly it also discusses potential applications to more challenging real-world cases and serve as the starting point for future further exploration into automated game play and beyond.

In vehicular networks (VNets), ensuring both road safety and dependable network connectivity is of utmost importance. Achieving this necessitates the creation of resilient and efficient decision-making policies that prioritize multiple objectives. In this paper, we develop a Variational Quantum Circuit (VQC)-based multi-objective reinforcement learning (MORL) framework to characterize efficient network selection and autonomous driving policies in a vehicular network (VNet). Numerical results showcase notable enhancements in both convergence rates and rewards when compared to conventional deep-Q networks (DQNs), validating the efficacy of the VQC-MORL solution.

For aligning large language models (LLMs), prior work has leveraged reinforcement learning via human feedback (RLHF) or variations of direct preference optimization (DPO). While DPO offers a simpler framework based on maximum likelihood estimation, it compromises on the ability to tune language models to easily maximize non-differentiable and non-binary objectives according to the LLM designer's preferences (e.g., using simpler language or minimizing specific kinds of harmful content). These may neither align with user preferences nor even be able to be captured tractably by binary preference data. To leverage the simplicity and performance of DPO with the generalizability of RL, we propose a hybrid approach between DPO and RLHF. With a simple augmentation to the implicit reward decomposition of DPO, we allow for tuning LLMs to maximize a set of arbitrary auxiliary rewards using offline RL. The proposed method, Hybrid Preference Optimization (HPO), shows the ability to effectively generalize to both user preferences and auxiliary designer objectives, while preserving alignment performance across a range of challenging benchmarks and model sizes.

Aligning human preference and value is an important requirement for contemporary foundation models. State-of-the-art techniques such as Reinforcement Learning from Human Feedback (RLHF) often consist of two stages: 1) supervised fine-tuning (SFT), where the model is fine-tuned by learning from human demonstration data; 2) Preference learning, where preference data is used to learn a reward model, which is in turn used by a reinforcement learning (RL) step to fine-tune the model. Such reward model serves as a proxy to human preference, and it is critical to guide the RL step towards improving the model quality. In this work, we argue that the SFT stage significantly benefits from learning a reward model as well. Instead of using the human demonstration data directly via supervised learning, we propose to leverage an Inverse Reinforcement Learning (IRL) technique to (explicitly or implicitly) build an reward model, while learning the policy model. This approach leads to new SFT algorithms that are not only efficient to implement, but also promote the ability to distinguish between the preferred and non-preferred continuations. Moreover, we identify a connection between the proposed IRL based approach, and certain self-play approach proposed recently, and showed that self-play is a special case of modeling a reward-learning agent. Theoretically, we show that the proposed algorithms converge to the stationary solutions of the IRL problem. Empirically, we align 1B and 7B models using proposed methods and evaluate them on a reward benchmark model and the HuggingFace Open LLM Leaderboard. The proposed methods show significant performance improvement over existing SFT approaches. Our results indicate that it is beneficial to explicitly or implicitly leverage reward learning throughout the entire alignment process.

Reinforcement learning is commonly concerned with problems of maximizing accumulated rewards in Markov decision processes. Oftentimes, a certain goal state or a subset of the state space attain maximal reward. In such a case, the environment may be considered solved when the goal is reached. Whereas numerous techniques, learning or non-learning based, exist for solving environments, doing so optimally is the biggest challenge. Say, one may choose a reward rate which penalizes the action effort. Reinforcement learning is currently among the most actively developed frameworks for solving environments optimally by virtue of maximizing accumulated reward, in other words, returns. Yet, tuning agents is a notoriously hard task as reported in a series of works. Our aim here is to help the agent learn a near-optimal policy efficiently while ensuring a goal reaching property of some basis policy that merely solves the environment. We suggest an algorithm, which is fairly flexible, and can be used to augment practically any agent as long as it comprises of a critic. A formal proof of a goal reaching property is provided. Simulation experiments on six problems under five agents, including the benchmarked one, provided an empirical evidence that the learning can indeed be boosted while ensuring goal reaching property.