cs.LG（2024-04-22）

📊 共 17 篇论文 | 🔗 3 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (11 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (4 🔗1) 支柱一：机器人控制 (Robot Control) (2)

🔬 支柱二：RL算法与架构 (RL & Architecture) (11 篇)

#题目一句话要点标签🔗⭐
1 Filtered Direct Preference Optimization 提出过滤直接偏好优化方法以提升文本质量 reinforcement learning RLHF DPO ✅
2 General Item Representation Learning for Cold-start Content Recommendations 提出一种通用项目表示学习框架以解决冷启动推荐问题 representation learning multimodal
3 Multi-view Disentanglement for Reinforcement Learning with Multiple Cameras 提出多视角解耦方法以解决图像基础强化学习中的相机依赖问题 reinforcement learning egocentric
4 Generalizing Multi-Step Inverse Models for Representation Learning to Finite-Memory POMDPs 提出通用多步逆模型以解决有限记忆POMDP中的状态表示学习问题 reinforcement learning representation learning
5 Preference Fine-Tuning of LLMs Should Leverage Suboptimal, On-Policy Data 提出基于次优在线数据的偏好微调方法以提升LLM性能 reinforcement learning contrastive learning large language model
6 Learning Control Barrier Functions and their application in Reinforcement Learning: A Survey 提出控制屏障函数以解决强化学习中的安全性问题 reinforcement learning
7 Brain-Inspired Continual Learning-Robust Feature Distillation and Re-Consolidation for Class Incremental Learning 提出鲁棒重演框架以解决持续学习中的灾难性遗忘问题 distillation
8 Deep Regression Representation Learning with Topology 提出PH-Reg以优化回归任务中的表示学习 representation learning ✅
9 FedTAD: Topology-aware Data-free Knowledge Distillation for Subgraph Federated Learning 提出FedTAD以解决子图联邦学习中的异构性问题 distillation
10 Distributional Black-Box Model Inversion Attack with Multi-Agent Reinforcement Learning 提出分布式黑箱模型反演攻击以解决隐私数据恢复问题 reinforcement learning
11 A survey of air combat behavior modeling using machine learning 通过机器学习建模空战行为以提升模拟训练效果 reinforcement learning imitation learning

🔬 支柱九：具身大模型 (Embodied Foundation Models) (4 篇)

#题目一句话要点标签🔗⭐
12 Mélange: Cost Efficient Large Language Model Serving by Exploiting GPU Heterogeneity 提出Mélange框架以降低大语言模型服务的GPU成本 large language model
13 Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing 提出混合推理方法以解决大模型部署成本与质量问题 large language model
14 An empirical study of LLaMA3 quantization: from LLMs to MLLMs 研究LLaMA3量化以解决低比特宽度下性能下降问题 large language model ✅
15 Towards smaller, faster decoder-only transformers: Architectural variants and their implications 提出三种解码器仅模型以解决模型体积与训练速度问题 large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
16 Multidimensional Adaptive Coefficient for Inference Trajectory Optimization in Flow and Diffusion 提出多维自适应系数以优化流动与扩散模型的推理轨迹 trajectory optimization
17 Lipschitz-Regularized Critics Lead to Policy Robustness Against Transition Dynamics Uncertainty 提出PPO-PGDLC以解决强化学习中的动态转移不确定性问题 locomotion reinforcement learning PPO

⬅️ 返回 cs.LG 首页 · 🏠 返回主页