-- We propose a decentralized reinforcement learning solution for multi-agent shepherding of non-cohesive targets using policy-gradient methods. This model-free framework effectively solves the shepherding problem without prior dynamics knowledge. Experiments demonstrate our method's effectiveness and scalability with increased target numbers and limited sensing capabilities. The shepherding problem in robotics exemplifies the problem of harnessing complex systems for control [1], [2]. It generally involves a group of actively controlled agents, termed herders, strategically influencing a group of passive agents, referred to as targets.

Recent estimates put the carbon footprint of Bitcoin and Ethereum at an average of 64 and 26 million tonnes of CO2 per year, respectively. To address this growing problem, several possible approaches have been proposed in the literature: creating alternative blockchain consensus mechanisms, applying redundancy reduction techniques, utilizing renewable energy sources, and employing energy-efficient devices, etc. In this paper, we follow the second avenue and propose an efficient approach based on reinforcement learning that improves the block broadcasting scheme in blockchain networks. The analysis and experimental results confirmed that the proposed improvement of the block propagation scheme could cleverly handle network dynamics and achieve better results than the default approach. Additionally, our technical integration of the simulator and developed RL environment can be used as a complete solution for further study of new schemes and protocols that use RL or other ML techniques.

Dynamic resource allocation in multi-agent settings often requires balancing efficiency with fairness over time--a challenge inadequately addressed by conventional, myopic fairness measures. Motivated by behavioral insights that human judgments of fairness evolve with temporal distance, we introduce a novel framework for temporal fairness that incorporates past-discounting mechanisms. By applying a tunable discount factor to historical utilities, our approach interpolates between instantaneous and perfect-recall fairness, thereby capturing both immediate outcomes and long-term equity considerations. Beyond aligning more closely with human perceptions of fairness, this past-discounting method ensures that the augmented state space remains bounded, significantly improving computational tractability in sequential decision-making settings. We detail the formulation of discounted-recall fairness in both additive and averaged utility contexts, illustrate its benefits through practical examples, and discuss its implications for designing balanced, scalable resource allocation strategies.

Many modern robotic systems such as multi-robot systems and manipulators exhibit redundancy, a property owing to which they are capable of executing multiple tasks. This work proposes a novel method, based on the Reinforcement Learning (RL) paradigm, to train redundant robots to be able to execute multiple tasks concurrently. Our approach differs from typical multi-objective RL methods insofar as the learned tasks can be combined and executed in possibly time-varying prioritized stacks. We do so by first defining a notion of task independence between learned value functions. We then use our definition of task independence to propose a cost functional that encourages a policy, based on an approximated value function, to accomplish its control objective while minimally interfering with the execution of higher priority tasks. This allows us to train a set of control policies that can be executed simultaneously. We also introduce a version of fitted value iteration to learn to approximate our proposed cost functional efficiently. We demonstrate our approach on several scenarios and robotic systems.

Recent advancements in Chain of Thought (COT) generation have significantly improved the reasoning capabilities of Large Language Models (LLMs), with reinforcement learning (RL) emerging as an effective post-training approach. Multimodal Large Language Models (MLLMs) inherit this reasoning potential but remain underexplored in tasks requiring both perception and logical reasoning. To address this, we introduce SEED-Bench-R1, a benchmark designed to systematically evaluate post-training methods for MLLMs in video understanding. It includes intricate real-world videos and complex everyday planning tasks in the format of multiple-choice questions, requiring sophisticated perception and reasoning. SEED-Bench-R1 assesses generalization through a three-level hierarchy: in-distribution, cross-environment, and cross-environment-task scenarios, equipped with a large-scale training dataset with easily verifiable ground-truth answers. Using Qwen2-VL-Instruct-7B as a base model, we compare RL with supervised fine-tuning (SFT), demonstrating RL's data efficiency and superior performance on both in-distribution and out-of-distribution tasks, even outperforming SFT on general video understanding benchmarks like LongVideoBench. Our detailed analysis reveals that RL enhances visual perception but often produces less logically coherent reasoning chains. We identify key limitations such as inconsistent reasoning and overlooked visual cues, and suggest future improvements in base model reasoning, reward modeling, and RL robustness against noisy signals.

We introduce Open-Reasoner-Zero, the first open source implementation of large-scale reasoning-oriented RL training focusing on scalability, simplicity and accessibility. Through extensive experiments, we demonstrate that a minimalist approach, vanilla PPO with GAE ($\lambda=1$, $\gamma=1$) and straightforward rule-based rewards, without any KL regularization, is sufficient to scale up both response length and benchmark performance, similar to the phenomenon observed in DeepSeek-R1-Zero. Using the same base model as DeepSeek-R1-Zero-Qwen-32B, our implementation achieves superior performance on AIME2024, MATH500, and the GPQA Diamond benchmark while demonstrating remarkable efficiency -- requiring only a tenth of the training steps, compared to DeepSeek-R1-Zero pipeline. In the spirit of open source, we release our source code, parameter settings, training data, and model weights across various sizes.

-- Domain randomization (DR) enables sim-to-real transfer by training controllers on a distribution of simulated environments, with the goal of achieving robust performance in the real world. Although DR is widely used in practice and is often solved using simple policy gradient (PG) methods, understanding of its theoretical guarantees remains limited. T oward addressing this gap, we provide the first convergence analysis of PG methods for domain-randomized linear quadratic regulation (LQR). We show that PG converges globally to the minimizer of a finite-sample approximation of the DR objective under suitable bounds on the heterogeneity of the sampled systems. We also quantify the sample-complexity associated with achieving a small performance gap between the sample-average and population-level objectives. Additionally, we propose and analyze a discount-factor annealing algorithm that obviates the need for an initial jointly stabilizing controller, which may be challenging to find. Empirical results support our theoretical findings and highlight promising directions for future work, including risk-sensitive DR formulations and stochastic PG algorithms. Domain randomization (DR) has emerged as a dominant paradigm to enable transfer of policies optimized in simulation to the real world by randomizing simulator parameters during training [1-3]. In doing so, just as with robust control, DR accounts for discrepancies between the model used in simulation to synthesize a policy and the system that it is deployed on. Since DR does not solely focus on optimizing the worst-case performance, it can result in less conservative controller performance while still ensuring robust stability with high probability. Furthermore, DR can be easily implemented via first order methods. This makes it straightforward to incorporate into a wide variety of reinforcement learning schemes and to benefit from the increasing availability of parallel computation. Despite the ease with which DR can be implemented using first order methods, ensuring convergence of these methods remains a critical challenge, with practitioners relying upon complex scheduling of various hyperparameters in the optimization procedure [3].

The rapid expansion of ride-sourcing services such as Uber, Lyft, and Didi Chuxing has fundamentally reshaped urban transportation by offering flexible, on-demand mobility via mobile applications. Despite their convenience, these platforms confront significant operational challenges, particularly vehicle rebalancing - the strategic repositioning of thousands of vehicles to address spatiotemporal mismatches in supply and demand. Inadequate rebalancing results in prolonged rider waiting times, inefficient vehicle utilization, and inequitable distribution of services, leading to disparities in driver availability and income. To tackle these complexities, we introduce scalable continuous-state mean-field control (MFC) and reinforcement learning (MFRL) models that explicitly represent each vehicle's precise location and employ continuous repositioning actions guided by the distribution of other vehicles. To ensure equitable service distribution, an accessibility constraint is integrated within our optimal control formulation, balancing operational efficiency with equitable access to the service across geographic regions. Our approach acknowledges realistic conditions, including inherent stochasticity in transitions, the simultaneous occurrence of vehicle-rider matching, vehicles' rebalancing and cruising, and variability in rider behaviors. Crucially, we relax the traditional mean-field assumption of equal supply-demand volume, better reflecting practical scenarios. Extensive empirical evaluation using real-world data-driven simulation of Shenzhen demonstrates the real-time efficiency and robustness of our approach at the scale of tens of thousands of vehicles. The code is available at https://github.com/mjusup1501/mf-vehicle-rebalancing.

We propose Rec-R1, a general reinforcement learning framework that bridges large language models (LLMs) with recommendation systems through closed-loop optimization. Unlike prompting and supervised fine-tuning (SFT), Rec-R1 directly optimizes LLM generation using feedback from a fixed black-box recommendation model, without relying on synthetic SFT data from proprietary models such as GPT-4o. This avoids the substantial cost and effort required for data distillation. To verify the effectiveness of Rec-R1, we evaluate it on two representative tasks: product search and sequential recommendation. Experimental results demonstrate that Rec-R1 not only consistently outperforms prompting- and SFT-based methods, but also achieves significant gains over strong discriminative baselines, even when used with simple retrievers such as BM25. Moreover, Rec-R1 preserves the general-purpose capabilities of the LLM, unlike SFT, which often impairs instruction-following and reasoning. These findings suggest Rec-R1 as a promising foundation for continual task-specific adaptation without catastrophic forgetting.

The economic feasibility of nuclear microreactors will depend on minimizing operating costs through advancements in autonomous control, especially when these microreactors are operating alongside other types of energy systems (e.g., renewable energy). This study explores the application of deep reinforcement learning (RL) for real-time drum control in microreactors, exploring performance in regard to load-following scenarios. By leveraging a point kinetics model with thermal and xenon feedback, we first establish a baseline using a single-output RL agent, then compare it against a traditional proportional-integral-derivative (PID) controller. This study demonstrates that RL controllers, including both single- and multi-agent RL (MARL) frameworks, can achieve similar or even superior load-following performance as traditional PID control across a range of load-following scenarios. In short transients, the RL agent was able to reduce the tracking error rate in comparison to PID. Over extended 300-minute load-following scenarios in which xenon feedback becomes a dominant factor, PID maintained better accuracy, but RL still remained within a 1% error margin despite being trained only on short-duration scenarios. This highlights RL's strong ability to generalize and extrapolate to longer, more complex transients, affording substantial reductions in training costs and reduced overfitting. Furthermore, when control was extended to multiple drums, MARL enabled independent drum control as well as maintained reactor symmetry constraints without sacrificing performance -- an objective that standard single-agent RL could not learn. We also found that, as increasing levels of Gaussian noise were added to the power measurements, the RL controllers were able to maintain lower error rates than PID, and to do so with less control effort.