cs.LG（2024-03-11）

📊 共 29 篇论文 | 🔗 7 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (15 🔗4) 支柱九：具身大模型 (Embodied Foundation Models) (9 🔗2) 支柱一：机器人控制 (Robot Control) (4 🔗1) 支柱八：物理动画 (Physics-based Animation) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (15 篇)

#题目一句话要点标签🔗⭐
1 Re-Simulation-based Self-Supervised Learning for Pre-Training Foundation Models 提出RS3L以解决物理科学中的自监督学习问题 representation learning contrastive learning foundation model
2 Tactical Decision Making for Autonomous Trucks by Deep Reinforcement Learning with Total Cost of Operation Based Reward 提出深度强化学习框架以优化自主卡车的战术决策 reinforcement learning deep reinforcement learning curriculum learning
3 Joint-Embedding Masked Autoencoder for Self-supervised Learning of Dynamic Functional Connectivity from the Human Brain 提出ST-JEMA以解决动态功能连接性学习中的标签稀缺问题 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture
4 Quantifying the Sensitivity of Inverse Reinforcement Learning to Misspecification 分析逆强化学习对行为模型误设的敏感性 reinforcement learning inverse reinforcement learning
5 Deep Reinforcement Learning for Modelling Protein Complexes 提出GAPN以解决多链蛋白复合体建模的挑战 reinforcement learning deep reinforcement learning
6 $\mathbf{(N,K)}$-Puzzle: A Cost-Efficient Testbed for Benchmarking Reinforcement Learning Algorithms in Generative Language Model 提出(N,K)-Puzzle以解决强化学习算法评估标准化问题 reinforcement learning PPO DPO
7 In-context Exploration-Exploitation for Reinforcement Learning 提出ICEE算法以提高离线强化学习的效率 reinforcement learning policy learning offline reinforcement learning
8 A multi-cohort study on prediction of acute brain dysfunction states using selective state space models 提出基于选择性状态空间模型的急性脑功能障碍预测方法 Mamba state space model
9 Acquiring Diverse Skills using Curriculum Reinforcement Learning with Mixture of Experts 提出Di-SkilL以解决强化学习中多样技能学习问题 reinforcement learning ✅
10 Probabilistic Contrastive Learning for Long-Tailed Visual Recognition 提出概率对比学习以解决长尾视觉识别问题 contrastive learning ✅
11 PeerAiD: Improving Adversarial Distillation from a Specialized Peer Tutor 提出PeerAiD以解决对抗蒸馏中的教师网络局限性问题 distillation ✅
12 AuG-KD: Anchor-Based Mixup Generation for Out-of-Domain Knowledge Distillation 提出AuG-KD以解决跨领域知识蒸馏中的知识转移问题 distillation ✅
13 A representation-learning game for classes of prediction tasks 提出基于博弈的表示学习方法以优化预测任务 representation learning
14 RL-MSA: a Reinforcement Learning-based Multi-line bus Scheduling Approach 提出基于强化学习的多线路公交调度方法以解决调度不确定性问题 reinforcement learning
15 ε-Neural Thompson Sampling of Deep Brain Stimulation for Parkinson Disease Treatment 提出ε-神经汤普森采样以优化帕金森病深脑刺激 reinforcement learning PULSE

🔬 支柱九：具身大模型 (Embodied Foundation Models) (9 篇)

#题目一句话要点标签🔗⭐
16 Process signature-driven high spatio-temporal resolution alignment of multimodal data 提出HiRA-Pro以解决多模态数据高时空分辨率对齐问题 multimodal TAMP
17 Impact of Noisy Supervision in Foundation Model Learning 提出NMTune以缓解基础模型学习中的标签噪声问题 foundation model
18 Smart-Infinity: Fast Large Language Model Training using Near-Storage Processing on a Real System 提出Smart-Infinity以解决存储带宽瓶颈问题 large language model
19 What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation 通过扰动视角提出量化方法以提升大语言模型效率 large language model
20 InfiBench: Evaluating the Question-Answering Capabilities of Code Large Language Models 提出InfiBench以评估代码大语言模型的问答能力 large language model ✅
21 SMART: Automatically Scaling Down Language Models with Accuracy Guarantees for Reduced Processing Fees 提出SMART框架以降低语言模型推理成本 large language model
22 Elephants Never Forget: Testing Language Models for Memorization of Tabular Data 提出多种技术评估LLM对表格数据的记忆能力 large language model ✅
23 ContextGPT: Infusing LLMs Knowledge into Neuro-Symbolic Activity Recognition Models 提出ContextGPT以解决人类活动识别中的数据稀缺问题 large language model
24 Unraveling the Mystery of Scaling Laws: Part I 提出新的缩放法则以优化大规模语言模型训练 large language model

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
25 Enhancing Joint Motion Prediction for Individuals with Limb Loss Through Model Reprogramming 通过模型重编程提升肢体缺失者的关节运动预测能力 manipulation motion prediction
26 DeepSafeMPC: Deep Learning-Based Model Predictive Control for Safe Multi-Agent Reinforcement Learning 提出DeepSafeMPC以解决安全多智能体强化学习问题 MPC model predictive control reinforcement learning
27 Unveiling the Significance of Toddler-Inspired Reward Transition in Goal-Oriented Reinforcement Learning 提出幼儿启发的奖励转变以提升目标导向强化学习效果 manipulation reinforcement learning egocentric
28 Can LLMs Separate Instructions From Data? And What Do We Even Mean By That? 提出指令与数据分离度量以解决LLM安全性问题 manipulation world model world models ✅

🔬 支柱八：物理动画 (Physics-based Animation) (1 篇)

#题目一句话要点标签🔗⭐
29 UPS: Efficiently Building Foundation Models for PDE Solving via Cross-Modal Adaptation 提出UPS以高效构建PDE求解的基础模型 spatiotemporal foundation model

⬅️ 返回 cs.LG 首页 · 🏠 返回主页