cs.LG（2024-12-30）

📊 共 10 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) (4 🔗2) 支柱九：具身大模型 (Embodied Foundation Models) (3 🔗1) 支柱八：物理动画 (Physics-based Animation) (2) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#	题目	一句话要点	标签	🔗	⭐
1	Frequency-Masked Embedding Inference: A Non-Contrastive Approach for Time Series Representation Learning	提出频率掩码嵌入推理(FEI)，用于解决时间序列表示学习中对比学习的局限性。	representation learning contrastive learning	✅
2	Prototypical Distillation and Debiased Tuning for Black-box Unsupervised Domain Adaptation	提出ProDDing框架，解决黑盒无监督域自适应中的知识迁移与偏差校正问题。	distillation	✅
3	An Unsupervised Anomaly Detection in Electricity Consumption Using Reinforcement Learning and Time Series Forest Based Framework	提出基于强化学习和时间序列森林的无监督异常检测模型选择框架，提升电力消耗异常检测性能。	reinforcement learning
4	LEASE: Offline Preference-based Reinforcement Learning with High Sample Efficiency	LEASE：高样本效率的离线偏好强化学习算法	reinforcement learning

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

#	题目	一句话要点	标签	🔗	⭐
5	TimeRAF: Retrieval-Augmented Foundation model for Zero-shot Time Series Forecasting	TimeRAF：检索增强的预训练模型用于零样本时间序列预测	foundation model
6	SafeSynthDP: Leveraging Large Language Models for Privacy-Preserving Synthetic Data Generation Using Differential Privacy	SafeSynthDP：利用大语言模型和差分隐私生成保护隐私的合成数据	large language model
7	Efficiently Scaling LLM Reasoning with Certaindex	提出Certaindex指标，通过提前退出加速LLM推理，提升服务吞吐量。	chain-of-thought	✅

🔬 支柱八：物理动画 (Physics-based Animation) (2 篇)

#	题目	一句话要点	标签	🔗	⭐
8	Conservation-informed Graph Learning for Spatiotemporal Dynamics Prediction	提出保守信息图学习以解决时空动态预测问题	spatiotemporal
9	Sparse chaos in cortical circuits	揭示神经元脉冲生成机制对皮层回路稀疏混沌状态的调控作用	PULSE

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#	题目	一句话要点	标签	🔗	⭐
10	A Novel Framework for Learning Stochastic Representations for Sequence Generation and Recognition	提出基于随机循环神经网络的参数偏置模型，用于序列生成与识别	motion generation

⬅️ 返回 cs.LG 首页 · 🏠 返回主页