cs.LG（2023-10-23）

📊 共 18 篇论文 | 🔗 2 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (8 🔗1) 支柱二：RL算法与架构 (RL & Architecture) (6) 支柱一：机器人控制 (Robot Control) (4 🔗1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (8 篇)

#题目一句话要点标签🔗⭐
1 Modality Dropout for Multimodal Device Directed Speech Detection using Verbal and Non-Verbal Features 提出多模态丢弃技术以提升设备导向语音检测性能 multimodal
2 Linear Representations of Sentiment in Large Language Models 揭示大型语言模型中情感的线性表示机制 large language model
3 Federated Learning of Large Language Models with Parameter-Efficient Prompt Tuning and Adaptive Optimization 提出FedPepTAO以解决大语言模型的联邦学习效率问题 large language model ✅
4 Multimodal Graph Learning for Modeling Emerging Pandemics with Big Data 提出MGL4MEP框架以解决新兴疫情预测问题 multimodal
5 DoGE: Domain Reweighting with Generalization Estimation 提出DoGE以优化大语言模型的领域重加权问题 large language model
6 SpecTr: Fast Speculative Decoding via Optimal Transport 提出SpecTr以加速自回归解码过程 large language model
7 Text2Topic: Multi-Label Text Classification System for Efficient Topic Detection in User Generated Content with Zero-Shot Capabilities 提出Text2Topic以解决多标签文本分类问题 large language model
8 FedSplitX: Federated Split Learning for Computationally-Constrained Heterogeneous Clients 提出FedSplitX以解决异构客户端的联邦分割学习问题 foundation model

🔬 支柱二：RL算法与架构 (RL & Architecture) (6 篇)

#题目一句话要点标签🔗⭐
9 Mind the Model, Not the Agent: The Primacy Bias in Model-based RL 提出世界模型重置以解决模型基础强化学习中的初始偏差问题 reinforcement learning world model world models
10 Corruption-Robust Offline Reinforcement Learning with General Function Approximation 提出一种抗干扰的离线强化学习算法以应对样本污染问题 reinforcement learning offline RL offline reinforcement learning
11 Diverse Priors for Deep Reinforcement Learning 提出多样性先验以解决深度强化学习中的不确定性问题 reinforcement learning deep reinforcement learning
12 A Doubly Robust Approach to Sparse Reinforcement Learning 提出双重稳健算法以解决稀疏强化学习问题 reinforcement learning
13 Graph Ranking Contrastive Learning: A Extremely Simple yet Efficient Method 提出GraphRank以解决图对比学习中的假负样本问题 contrastive learning
14 Making RL with Preference-based Feedback Efficient via Randomization 提出基于随机化的RLHF算法以提高样本效率 reinforcement learning RLHF

🔬 支柱一：机器人控制 (Robot Control) (4 篇)

#题目一句话要点标签🔗⭐
15 Iteratively Learn Diverse Strategies with State Distance Information 提出基于状态距离信息的多样性驱动强化学习算法解决策略多样性问题 locomotion reinforcement learning
16 SimBIG: Field-level Simulation-Based Inference of Galaxy Clustering 提出SimBIG框架以解决宇宙学参数推断问题 MPC
17 Harnessing Attention Mechanisms: Efficient Sequence Reduction using Attention-based Autoencoders 提出基于注意力机制的自编码器以高效减少序列长度 manipulation
18 Inferring Relational Potentials in Interacting Systems 提出NIIP方法以推断交互系统中的关系潜力 manipulation ✅

⬅️ 返回 cs.LG 首页 · 🏠 返回主页