cs.LG（2024-02-20）

📊 共 20 篇论文 | 🔗 2 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) (7) 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 Analysis of Using Sigmoid Loss for Contrastive Learning 提出双常数嵌入模型以优化对比学习中的Sigmoid损失 contrastive learning
2 Discovering Behavioral Modes in Deep Reinforcement Learning Policies Using Trajectory Clustering in Latent Space 提出基于潜在空间轨迹聚类的新方法以理解深度强化学习行为模式 reinforcement learning deep reinforcement learning DRL
3 Align Your Intents: Offline Imitation Learning via Optimal Transport 提出AILOT方法以解决离线模仿学习中的奖励缺失问题 reinforcement learning offline RL offline reinforcement learning
4 An Equivariant Pretrained Transformer for Unified 3D Molecular Representation Learning 提出等变预训练变换器以解决跨领域3D分子表示学习问题 representation learning foundation model
5 Bayesian Reward Models for LLM Alignment 提出贝叶斯奖励模型以解决LLM对齐中的奖励过度优化问题 reinforcement learning RLHF large language model
6 SPRINQL: Sub-optimal Demonstrations driven Offline Imitation Learning 提出SPRINQL以解决离线模仿学习中的专家示范不足问题 imitation learning behavior cloning ✅
7 Improve Cross-Architecture Generalization on Dataset Distillation 提出模型池方法以改善数据集蒸馏的跨架构泛化能力 distillation
8 Statistical curriculum learning: An elimination algorithm achieving an oracle risk 提出统计课程学习方法以实现最佳风险估计 curriculum learning
9 FGAD: Self-boosted Knowledge Distillation for An Effective Federated Graph Anomaly Detection Framework 提出FGAD框架以解决图异常检测中的隐私和非IID问题 distillation
10 Uniform Last-Iterate Guarantee for Bandits and Reinforcement Learning 提出统一最后迭代保证以解决强化学习中的即时性能问题 reinforcement learning
11 Reflect-RL: Two-Player Online RL Fine-Tuning for LMs 提出Reflect-RL以解决语言模型在线强化学习微调问题 reinforcement learning curriculum learning ✅

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
12 Chain of Thought Empowers Transformers to Solve Inherently Serial Problems 提出链式思维以增强变换器解决串行问题的能力 large language model chain-of-thought
13 Harnessing Large Language Models as Post-hoc Correctors 提出LlmCorr框架以低成本提升机器学习模型性能 large language model
14 APT-MMF: An advanced persistent threat actor attribution method based on multimodal and multilevel feature fusion 提出APT-MMF以解决APT行为者归属问题 multimodal
15 Thermometer: Towards Universal Calibration for Large Language Models 提出THERMOMETER以解决大语言模型的校准问题 large language model
16 TorchCP: A Python Library for Conformal Prediction 提出TorchCP以解决深度学习中的不确定性量化问题 large language model
17 Defending Jailbreak Prompts via In-Context Adversarial Game 提出ICAG以解决大型语言模型的越狱攻击问题 large language model
18 TRAP: Targeted Random Adversarial Prompt Honeypot for Black-Box Identification 提出TRAP以解决黑箱身份验证问题 large language model

🔬 支柱一：机器人控制 (Robot Control) (2 篇)

#题目一句话要点标签🔗⭐
19 FIDLAR: Forecast-Informed Deep Learning Architecture for Flood Mitigation 提出FIDLAR以优化洪水管理问题 MPC model predictive control
20 Byzantine-Robust Federated Learning: Impact of Client Subsampling and Local Updates 提出鲁棒的联邦学习方法以应对拜占庭客户端问题 manipulation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页