Update README.md

README.md

@@ -4,7 +4,7 @@
 
 ***We expose this user-friendly platform for beginners who intend to start federated learning (FL) study.***
 
-- *Now there are **30 FL/pFL algorithms, 3 scenarios, and 14 datasets** in this platform.*
+- *Now there are **30 traditional FL (tFL) or personalized FL (pFL) algorithms, 3 scenarios, and 14 datasets** in this platform.*
 
 - *This platform can simulate scenarios (e.g., Cifar100) with **more than 500 clients** using **a single GPU card** (e.g., 2080Ti, 11GB memory).*
 
@@ -15,33 +15,33 @@ Due to the frequent update, please download the **master branch** as the latest
 
 The origin of the **statistical heterogeneity** phenomenon is the personalization of users, who generate the non-IID (not Independent and Identically Distributed) and unbalanced data. With statistical heterogeneity existing in the FL scenario, a myriad of approaches have been proposed to crack this hard nut. In contrast, the personalized FL (pFL) may take the advantage of the statistically heterogeneious data to learn the personalized model for each user. 
 
-Thanks to [@Stonesjtu](https://github.com/Stonesjtu/pytorch_memlab/blob/d590c489236ee25d157ff60ecd18433e8f9acbe3/pytorch_memlab/mem_reporter.py#L185), this platform can also record the **GPU memory usage** for the model. By using the package [opacus](https://opacus.ai/), we introduce **DP (differential privacy)** into this platform (please refer to `./system/flcore/clients/clientavg.py` for example). Following [FedCG](https://www.ijcai.org/proceedings/2022/0324), we also introduce the **[DLG (Deep Leakage from Gradients)](https://papers.nips.cc/paper_files/paper/2019/hash/60a6c4002cc7b29142def8871531281a-Abstract.html) attack** and **PSNR (Peak Signal-to-Noise Ratio) metric** to evaluate the privacy-preserving ability of FL/pFL algorithms (please refer to `./system/flcore/servers/serveravg.py` for example). *Now we can train on some clients and evaluate on other new clients by setting `args.num_new_clients` in `./system/main.py`. Note that not all the FL/pFL algorithms support this feature.*
+Thanks to [@Stonesjtu](https://github.com/Stonesjtu/pytorch_memlab/blob/d590c489236ee25d157ff60ecd18433e8f9acbe3/pytorch_memlab/mem_reporter.py#L185), this platform can also record the **GPU memory usage** for the model. By using the package [opacus](https://opacus.ai/), we introduce **DP (differential privacy)** into this platform (please refer to `./system/flcore/clients/clientavg.py` for example). Following [FedCG](https://www.ijcai.org/proceedings/2022/0324), we also introduce the **[DLG (Deep Leakage from Gradients)](https://papers.nips.cc/paper_files/paper/2019/hash/60a6c4002cc7b29142def8871531281a-Abstract.html) attack** and **PSNR (Peak Signal-to-Noise Ratio) metric** to evaluate the privacy-preserving ability of tFL/pFL algorithms (please refer to `./system/flcore/servers/serveravg.py` for example). *Now we can train on some clients and evaluate on other new clients by setting `args.num_new_clients` in `./system/main.py`. Note that not all the tFL/pFL algorithms support this feature.*
 
 
 ## Algorithms with code (updating)
 
-> ### Traditional FL
+> ### Traditional FL (tFL)
 
 - **FedAvg** — [Communication-Efficient Learning of Deep Networks from Decentralized Data](http://proceedings.mlr.press/v54/mcmahan17a.html) *AISTATS 2017*
 
-  ***Update-correction-based FL***
+  ***Update-correction-based tFL***
 
 - **SCAFFOLD** - [SCAFFOLD: Stochastic Controlled Averaging for Federated Learning](http://proceedings.mlr.press/v119/karimireddy20a.html) *ICML 2020*
 
-  ***Regularization-based FL***
+  ***Regularization-based tFL***
 
 - **FedProx** — [Federated Optimization in Heterogeneous Networks](https://proceedings.mlsys.org/paper/2020/hash/38af86134b65d0f10fe33d30dd76442e-Abstract.html) *MLsys 2020*
 - **FedDyn** — [Federated Learning Based on Dynamic Regularization](https://openreview.net/forum?id=B7v4QMR6Z9w) *ICLR 2021*
 
-  ***Model-splitting-based FL***
+  ***Model-splitting-based tFL***
 
 - **MOON** — [Model-Contrastive Federated Learning](https://openaccess.thecvf.com/content/CVPR2021/html/Li_Model-Contrastive_Federated_Learning_CVPR_2021_paper.html) *CVPR 2021*
 
-  ***Knowledge-distillation-based FL***
+  ***Knowledge-distillation-based tFL***
 
 - **FedGen** — [Data-Free Knowledge Distillation for Heterogeneous Federated Learning](http://proceedings.mlr.press/v139/zhu21b.html) *ICML 2021*
 
-> ### Personalized FL
+> ### Personalized FL (pFL)
 
 - **FedMTL (not MOCHA)** — [Federated multi-task learning](https://papers.nips.cc/paper/2017/hash/6211080fa89981f66b1a0c9d55c61d0f-Abstract.html) *NeurIPS 2017*
 - **FedBN** — [FedBN: Federated Learning on non-IID Features via Local Batch Normalization](https://openreview.net/forum?id=6YEQUn0QICG) *ICLR 2021*