cs.LG（2024-08-24）

📊 共 8 篇论文 | 🔗 1 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (5) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (5 篇)

#题目一句话要点标签🔗⭐
1 Hybrid Training for Enhanced Multi-task Generalization in Multi-agent Reinforcement Learning 提出HyGen框架以解决多智能体强化学习中的多任务泛化问题 reinforcement learning
2 Rethinking State Disentanglement in Causal Reinforcement Learning 在因果强化学习中，重新思考状态解耦问题，提出更宽松约束的解耦方法。 reinforcement learning
3 Thresholded Lexicographic Ordered Multiobjective Reinforcement Learning 提出阈值化词典序多目标强化学习算法，解决现有方法理论不足和实践问题。 reinforcement learning
4 Disentangled Generative Graph Representation Learning 提出DiGGR：解耦生成图表示学习框架，提升图表示的鲁棒性和可解释性。 representation learning
5 Reinforcement Learning for Causal Discovery without Acyclicity Constraints ALIAS：一种无环约束的强化学习因果发现方法 reinforcement learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (3 篇)

#题目一句话要点标签🔗⭐
6 A Law of Next-Token Prediction in Large Language Models 揭示大语言模型中token预测的普适定律：层间贡献均等 large language model
7 LlamaDuo: LLMOps Pipeline for Seamless Migration from Service LLMs to Small-Scale Local LLMs LlamaDuo：LLMOps流水线，实现服务LLM到小型本地LLM的无缝迁移 large language model ✅
8 DOPPLER: Differentially Private Optimizers with Low-pass Filter for Privacy Noise Reduction 提出DOPPLER：一种基于低通滤波的差分隐私优化器，用于降低隐私噪声的影响。 foundation model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页