Adaptive Bayesian Partner Selection for Federated Clinical Centers
为解决医疗联邦学习中的异质性和时间概念漂移问题,提出自适应贝叶斯伙伴选择(ABPS)框架,通过轻量级机制优化合作,减少通信成本。
为解决医疗联邦学习中的异质性和时间概念漂移问题,提出自适应贝叶斯伙伴选择(ABPS)框架,通过轻量级机制优化合作,减少通信成本。
该研究提出了一种多模态-多分辨率基础模型,用于月球遥感,通过在SomBench数据集上预训练来解决多种任务如陨石坑检测、不规则海斑分割和极地冰探测回归。
为解决月球科学中机器学习的可重复性问题,创建了包含30多个层的统一数据集SomBench,并通过ResNet-50和SwinV2-B模型验证了其有效性。
为解决LEO星座与射电天文服务之间的频谱干扰问题,提出SkyShare系统,通过预测性波束调度和优化算法,在保护天文观测的同时保持网络覆盖。
This work addresses the limitations of existing personalized federated reinforcement learning methods, which overly rely on extrinsic rewards and suffer from insufficient exploration in non-stationary or sparse-reward environments, leading to weak policy personalization and low sample efficiency. To overcome these challenges, we propose the first exploration-driven personalized federated reinforcement learning framework that integrates intrinsic motivation. Specifically, clients leverage Random Network Distillation (RND) to generate intrinsic rewards that enhance local exploration, while the server aggregates only minimal novelty summaries and broadcasts a global exploration prior to coordinate diverse cross-client exploration without compromising privacy. Experimental results demonstrate that our approach significantly outperforms state-of-the-art methods on standard benchmarks as well as in sparse and delayed reward settings, achieving both stronger policy personalization and higher sample efficiency.
为解决医疗联邦学习中的异质性和时间概念漂移问题,提出自适应贝叶斯伙伴选择(ABPS)框架,通过轻量级机制优化合作,减少通信成本。
该研究提出了一种多模态-多分辨率基础模型,用于月球遥感,通过在SomBench数据集上预训练来解决多种任务如陨石坑检测、不规则海斑分割和极地冰探测回归。
为解决月球科学中机器学习的可重复性问题,创建了包含30多个层的统一数据集SomBench,并通过ResNet-50和SwinV2-B模型验证了其有效性。
为解决LEO星座与射电天文服务之间的频谱干扰问题,提出SkyShare系统,通过预测性波束调度和优化算法,在保护天文观测的同时保持网络覆盖。
This work addresses the limitations of existing personalized federated reinforcement learning methods, which overly rely on extrinsic rewards and suffer from insufficient exploration in non-stationary or sparse-reward environments, leading to weak policy personalization and low sample efficiency. To overcome these challenges, we propose the first exploration-driven personalized federated reinforcement learning framework that integrates intrinsic motivation. Specifically, clients leverage Random Network Distillation (RND) to generate intrinsic rewards that enhance local exploration, while the server aggregates only minimal novelty summaries and broadcasts a global exploration prior to coordinate diverse cross-client exploration without compromising privacy. Experimental results demonstrate that our approach significantly outperforms state-of-the-art methods on standard benchmarks as well as in sparse and delayed reward settings, achieving both stronger policy personalization and higher sample efficiency.