Tootfinder

Online Training and Pruning of Deep Reinforcement Learning Networks
Valentin Frank Ingmar Guenter, Athanasios Sideris
https://arxiv.org/abs/2507.11975 http…

Online Training and Pruning of Deep Reinforcement Learning Networks
Scaling deep neural networks (NN) of reinforcement learning (RL) algorithms has been shown to enhance performance when feature extraction networks are used but the gained performance comes at the significant expense of increased computational and memory complexity. Neural network pruning methods have successfully addressed this challenge in supervised learning. However, their application to RL is underexplored. We propose an approach to integrate simultaneous training and pruning within advance…

Dynamic Preference Multi-Objective Reinforcement Learning for Internet Network Management
DongNyeong Heo, Daniela Noemi Rim, Heeyoul Choi
https://arxiv.org/abs/2506.13153

Dynamic Preference Multi-Objective Reinforcement Learning for Internet Network Management
An internet network service provider manages its network with multiple objectives, such as high quality of service (QoS) and minimum computing resource usage. To achieve these objectives, a reinforcement learning-based (RL) algorithm has been proposed to train its network management agent. Usually, their algorithms optimize their agents with respect to a single static reward formulation consisting of multiple objectives with fixed importance factors, which we call preferences. However, in pract…

Efficient Preparation of Fermionic Superfluids in an Optical Dipole Trap through Reinforcement Learning
Yueyang Min, Ziliang Li, Yi Zhong, Jia-An Xuan, Jian Lin, Fei Leng, Xiaopeng Li
https://arxiv.org/abs/2507.12152

Efficient Preparation of Fermionic Superfluids in an Optical Dipole Trap through Reinforcement Learning
We demonstrate a reinforcement learning (RL) based control framework for optimizing evaporative cooling in the preparation of strongly interacting degenerate Fermi gases of Li6. Using a Soft Actor-Critic (SAC) algorithm, the system autonomously explores a high-dimensional parameter space to learn optimal cooling trajectories. Compared to conventional exponential ramps, our method achieves up to 130% improvement in atomic density within a 0.5 second, revealing non-trivial control strategies that…

Wasserstein-Barycenter Consensus for Cooperative Multi-Agent Reinforcement Learning
Ali Baheri
https://arxiv.org/abs/2506.12497 https://

Wasserstein-Barycenter Consensus for Cooperative Multi-Agent Reinforcement Learning
Cooperative multi-agent reinforcement learning (MARL) demands principled mechanisms to align heterogeneous policies while preserving the capacity for specialized behavior. We introduce a novel consensus framework that defines the team strategy as the entropic-regularized $p$-Wasserstein barycenter of agents' joint state--action visitation measures. By augmenting each agent's policy objective with a soft penalty proportional to its Sinkhorn divergence from this barycenter, the proposed approach …

Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice Theory
Jiancong Xiao, Zhekun Shi, Kaizhao Liu, Qi Long, Weijie J. Su
https://arxiv.org/abs/2506.12350

Theoretical Tensions in RLHF: Reconciling Empirical Success with Inconsistencies in Social Choice Theory
Despite its empirical success, Reinforcement Learning from Human Feedback (RLHF) has been shown to violate almost all the fundamental axioms in social choice theory -- such as majority consistency, pairwise majority consistency, and Condorcet consistency. This raises a foundational question: why does RLHF perform so well in practice if it fails these seemingly essential properties? In this paper, we resolve this paradox by showing that under mild and empirically plausible assumptions on the pre…

Analytical coarse grained potential parameterization by Reinforcement Learning for anisotropic cellulose
Xu Don
https://arxiv.org/abs/2506.12893 https://…

Analytical coarse grained potential parameterization by Reinforcement Learning for anisotropic cellulose
Cellulose nanocrystals(CNC) is a nano form of cellulose with great mechanical performance and other merit attributes. Revealed by preceding researches of experiments and molecular simulations, CNC has an anisotropic structure, in which hydrogen bonds play a pivotal role. Understanding structure and mechanical behavior of CNC in a mesoscopic scale is critical to improve manufacturing process and property of final products, in which experiment observation and atomistic simulation are not appropri…

Vane rheology of a fiber-reinforced granular material
Ladislas Wierzchalek, Georges Gauthier, Baptiste Darbois-Texier
https://arxiv.org/abs/2506.11762 http…

Vane rheology of a fiber-reinforced granular material
The addition of a small quantity of flexible fibers in a granular material is an efficient technique to increase the yield stress of the material. While the influence of fiber addition on the mechanical strength of granular media has been studied, much less is known about the flow properties of grain-fiber mixtures. In this article, we explore the effect of flexible fibers on the flow behavior of grain-fiber mixtures above the yield stress. We use a vane geometry to study the rheology of a dry …

Learning to Tune Like an Expert: Interpretable and Scene-Aware Navigation via MLLM Reasoning and CVAE-Based Adaptation
Yanbo Wang, Zipeng Fang, Lei Zhao, Weidong Chen
https://arxiv.org/abs/2507.11001

Learning to Tune Like an Expert: Interpretable and Scene-Aware Navigation via MLLM Reasoning and CVAE-Based Adaptation
Service robots are increasingly deployed in diverse and dynamic environments, where both physical layouts and social contexts change over time and across locations. In these unstructured settings, conventional navigation systems that rely on fixed parameters often fail to generalize across scenarios, resulting in degraded performance and reduced social acceptance. Although recent approaches have leveraged reinforcement learning to enhance traditional planners, these methods often fail in real-w…

Trust-MARL: Trust-Based Multi-Agent Reinforcement Learning Framework for Cooperative On-Ramp Merging Control in Heterogeneous Traffic Flow
Jie Pan, Tianyi Wang, Christian Claudel, Jing Shi
https://arxiv.org/abs/2506.12600

Trust-MARL: Trust-Based Multi-Agent Reinforcement Learning Framework for Cooperative On-Ramp Merging Control in Heterogeneous Traffic Flow
Intelligent transportation systems require connected and automated vehicles (CAVs) to conduct safe and efficient cooperation with human-driven vehicles (HVs) in complex real-world traffic environments. However, the inherent unpredictability of human behaviour, especially at bottlenecks such as highway on-ramp merging areas, often disrupts traffic flow and compromises system performance. To address the challenge of cooperative on-ramp merging in heterogeneous traffic environments, this study pro…

Kevin: Multi-Turn RL for Generating CUDA Kernels
Carlo Baronio, Pietro Marsella, Ben Pan, Simon Guo, Silas Alberti
https://arxiv.org/abs/2507.11948 https:/…

Kevin: Multi-Turn RL for Generating CUDA Kernels
Writing GPU kernels is a challenging task and critical for AI systems' efficiency. It is also highly iterative: domain experts write code and improve performance through execution feedback. Moreover, it presents verifiable rewards like correctness and speedup, making it a natural environment to apply Reinforcement Learning (RL). To explicitly incorporate the iterative nature of this process into training, we develop a flexible multi-turn RL recipe that addresses unique challenges encountered in…