| 1 |
Q-value Regularized Decision ConvFormer for Offline Reinforcement Learning |
提出Q值正则化Decision ConvFormer，提升离线强化学习轨迹拼接能力 |
reinforcement learning offline RL offline reinforcement learning |
|
|
| 2 |
Tera-SpaceCom: GNN-based Deep Reinforcement Learning for Joint Resource Allocation and Task Offloading in TeraHertz Band Space Networks |
提出基于GNN-DRL的GRANT算法，解决太赫兹空间网络中联合资源分配和任务卸载问题 |
reinforcement learning deep reinforcement learning DRL |
|
|
| 3 |
Scores as Actions: a framework of fine-tuning diffusion models by continuous-time reinforcement learning |
提出基于连续时间强化学习的扩散模型微调框架，提升生成质量 |
reinforcement learning RLHF |
|
|
| 4 |
GRE^2-MDCL: Graph Representation Embedding Enhanced via Multidimensional Contrastive Learning |
提出GRE^2-MDCL模型，通过多维对比学习增强图表示嵌入，提升节点分类性能。 |
representation learning contrastive learning |
|
|
| 5 |
Theoretical guarantees in KL for Diffusion Flow Matching |
提出Diffusion Flow Matching以解决生成模型的KL散度问题 |
flow matching |
|
|
| 6 |
Multiplex Graph Contrastive Learning with Soft Negatives |
提出MUX-GCL，利用多重图表示和软负样本进行跨尺度图对比学习。 |
contrastive learning |
✅ |
|
| 7 |
DiReDi: Distillation and Reverse Distillation for AIoT Applications |
提出DiReDi框架，通过知识蒸馏与反向蒸馏实现AIoT边缘模型自适应更新与用户隐私保护。 |
distillation |
|
|
| 8 |
DFDG: Data-Free Dual-Generator Adversarial Distillation for One-Shot Federated Learning |
提出DFDG：一种用于单次联邦学习的无数据双生成器对抗蒸馏方法 |
distillation |
|
|
| 9 |
Multi-Model based Federated Learning Against Model Poisoning Attack: A Deep Learning Based Model Selection for MEC Systems |
提出基于多模型的联邦学习框架，增强模型投毒攻击的防御能力，并应用于MEC系统。 |
reinforcement learning deep reinforcement learning |
|
|
| 10 |
Learning Causally Invariant Reward Functions from Diverse Demonstrations |
提出基于因果不变性的逆强化学习正则化方法，提升奖励函数泛化性 |
reinforcement learning inverse reinforcement learning |
|
|