# OVERVIEW of paper about NON-IID in FL

1. **NON-IID types for FL and NON-IID solution survey**
* [ZXL+21][Federated Learning on Non-IID Data: A Survey](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rkxtkCFN6>)
* [MZL+22][A state-of-the-art survey on solving non-IID data in Federated Learning](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/Hk5d8bsNp>)
* [LDC+21][Federated Learning on Non-IID Data Silos: An Experimental Study](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/Hyp5hWjN6>)
* [ZLL+22][Federated Learning with Non-IID Data](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/HkTey-iET>)　 <font color="#f00">important</font>
2. **FL client selection mechanism with NON-IID**
1. **RL**
* [AUCTION: Automated and Quality-Aware Client Selection Framework for Efficient Federated Learning](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/S14RrMsVp>)
* [Optimizing Federated Learning on Non-IID Data ](<http://example.com/>) (not record) <font color="#f00">important</font>
* [Optimizing Federated Learning With Deep Reinforcement Learning for Digital Twin Empowered　Industrial IoT　 ](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/Hy8HcGiEa>)
3. **stackberg + RL**
* [Make Smart Decisions Faster: Deciding D2D Resource Allocation via Stackelberg Game Guided Multi- Agent Deep Reinforcement Learning](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/HJhmO-o46>)
4. **others**
* [Client Selection for Federated Learning With Non-IID Data in Mobile Edge Computing](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rJvsJGjNT>)
3. **NON-IID aggregation by determine weight**
1. **RL**
* [FedDRL: Deep Reinforcement Learning-based Adaptive Aggregation for Non-IID Data in Federated Learning](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/BJKsy3FSp>)
* [ZLZ+23][Joint Device Scheduling and Bandwidth Allocation for Federated Learning over Wireless Networks](<https://>) <font color="#f00">再上傳筆記</font>
* [ZLT+22][Deep reinforcement learning based scheduling strategy for federated learning in sensor-cloud systems](<https://>) (PPO) <font color="#f00">再上傳筆記</font>
2. **Optimization**
* [Personalized Cross-Silo Federated Learning on Non-IID Data](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/BJb99ljrT>)
4. **FL client selection mechanism wit IID**
1. **RL**
* [Applicability of Deep Reinforcement Learning for Efficient](<http://example.com/>) (not finish study) <font color="#f00">important</font>
* [Optimizing Federated Learning in Distributed Industrial IoT: A Multi-Agent Approach Federated　Learning in Massive IoT Communications ](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/H1Zfi-oNT>)
* [Trust‑Augmented Deep Reinforcement Learning for Federated Learning Client Selection](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rydV5HuB6>) (DQN)
* [Experience-Driven Computational Resource Allocation of Federated Learning by Deep Reinforcement Learning](<https://>) (PPO) <font color="#f00">再上傳筆記</font>
* [Resource Allocation in Mobility-Aware Federated Learning Networks: A Deep Reinforcement Learning Approach](<https://>) (DQN) <font color="#f00">再上傳筆記</font>
* [Deep reinforcement learning based efficient access scheduling algorithm with an adaptive number of devices for federated learning IoT systems](<https://>) (ppo) <font color="#f00">未整理</font>
2. **matching**
* [Matching-Theory-Based Low-Latency Scheme for Multitask Federated Learning in MEC Networks](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/BkX7yZvST>) (multi-task)
3. **coalition**
* [Coalition based utility and efficiency optimization for multi-task federated learning in Internet of Vehicles](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/Bk7UvB7Lp>)
5. **Multimodel FL (MFL)**
* [Multimodal Federated Learning: A Survey](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rk9tWzj46>)
6. **Personalised FL**
1. network design
* [Multi-Task Federated Learning for Personalised
Deep Neural Networks in Edge Computing](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rky2lRSBa>)

7. **cluster FL**(之後可以再歸類到更詳細，基本上是)
* [Federated learning with hierarchical clustering of local updates to improve training on non-IID data](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/rk8gWMUrp>)
* [Personalized Federated Learning Algorithm with Adaptive Clustering for Non-IID IoT Data Incorporating Multi-Task Learning and Neural Network Model Characteristics](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/HyAiFq_ST>)
8. **multitask FL**
* [Joint Client Selection and Task Assignment for Multi-Task Federated Learning in MEC Networks](<https://>) <font color="#f00">再上傳筆記</font>
* [Bandwidth Allocation for Multiple Federated Learning Services in Wireless Edge Networks](<https://>) <font color="#f00">再上傳筆記</font>
* [Toward Multiple Federated Learning Services Resource Sharing in Mobile Edge Networks](<https://>) <font color="#f00">再上傳筆記</font>
* [Multi-Task Network Anomaly Detection using Federated Learning](<https://>) <font color="#f00">再上傳筆記</font>
* [Multi-Model Federated Learning](<https://>) <font color="#f00">再上傳筆記</font>
10. **others**
1. **resourece allocation for model aggregation in multitask FL**
* [A Federated Learning Multi-Task Scheduling Mechanism Based on Trusted Computing Sandbox](<https://hackmd.io/@nnlijpg3Ts6jTgdB-pUfEw/HJJLVFKrT>)

# 還沒分類
1. [DRL-Based Joint Resource Allocation and Device Orchestration for Hierarchical Federated Learning in NOMA-Enabled Industrial IoT](<https://hackmd.io/9LoPBduiQGS0BOH-XXUeyg?view>)
* RL,同時考慮到computation以及communication的resource allocation
2. [Exploring Deep-Reinforcement-Learning-Assisted Federated Learning for Online Resource Allocation in Privacy-Preserving EdgeIoT](<https://hackmd.io/DjGdL-goTM-beIzOe2gb9Q>) ~ 再整理清楚一些
* 用RL TD3
* 考慮到battery
* action space超級大

other:
EMD怎麼算(非常清楚): https://www.hmoonotes.org/2020/06/earth-movers-distance.html

[XW+21][Client Selection and Bandwidth Allocation in Wireless Federated Learning Networks: A Long-Term Perspective](<https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=9237168>)