update for pFL paper list (#561)

materials/paper_list/Personalized_FL/README.md

@@ -1,6 +1,16 @@
 ## Personalized Federated Learning
 This list is constantly being updated. Feel free to contribute!
 
+### 2023
+| Title | Venue | Link | Keywords | Note |
+| --- | --- | --- | --- | --- |
+| Personalized Federated Learning with Feature Alignment and Classifier Collaboration | ICLR | [pdf](https://openreview.net/pdf?id=SXZr8aDKia) | Collaboration | feature alignment by regularization, theoretically-guaranteed heads combination
+| Test-Time Robust Personalization for Federated Learning | ICLR | [pdf](https://openreview.net/pdf?id=3aBuJEza5sq) | Test-time Robustness |
+| A Statistical Framework for Personalized Federated Learning and Estimation: Theory, Algorithms, and Privacy | ICLR | [pdf](https://openreview.net/pdf?id=FUiDMCr_W4o) | Statistical Estimation, Differential Privacy, Empirical/Hierarchical Bayes |
+| FiT: Parameter Efficient Few-shot Transfer Learning for Personalized and Federated Image Classification | ICLR | [pdf](https://openreview.net/pdf?id=9aokcgBVIj1) | few-shot learning, transfer learning |
+| PerFedMask: Personalized Federated Learning with Optimized Masking Vectors | ICLR | [pdf](https://openreview.net/pdf?id=hxEIgUXLFF) | Masking vectors |
+| The Best of Both Worlds: Accurate Global and Personalized Models through Federated Learning with Data-Free Hyper-Knowledge Distillation | ICLR | [pdf](https://openreview.net/pdf?id=29V3AWjVAFi) | Knowledge Distillation, Differential Privacy, | share means of local data representations and soft predictions; no public data
+
 ### 2022
 | Title | Venue | Link | Keywords | Note |
 | --- | --- | --- | --- | --- |
@@ -16,18 +26,18 @@ This list is constantly being updated. Feel free to contribute!
 | Personalized Federated Learning via Variational Bayesian Inference | ICML | [pdf](https://arxiv.org/pdf/2206.07977.pdf) |  Bayesian variational inference; Upper bound |
 | Federated Learning with Partial Model Personalization | ICML | [pdf](https://proceedings.mlr.press/v162/pillutla22a/pillutla22a.pdf) | Partial model parameters; Transformer | 
 | On Bridging Generic and Personalized Federated Learning for Image Classification | ICLR | [pdf](https://arxiv.org/pdf/2107.00778) |  Partial model parameters; |
-| FedBABU: Toward Enhanced Representation for Federated Image Classification | ICLR | [pdf](https://openreview.net/pdf?id=HuaYQfggn5u)  | Partial model parameters; |
+| FedBABU: Toward Enhanced Representation for Federated Image Classification | ICLR | [pdf](https://openreview.net/pdf?id=HuaYQfggn5u)  | Partial model parameters; | keep the head (classifer) unchanged during FL training, then conduct fine-tuning before inference
 | Towards Personalized Federated Learning | Transactions on Neural Networks and Learning Systems | [pdf](https://arxiv.org/pdf/2103.00710)| Survey |
 
 ### 2021
 | Title | Venue | Link | Keywords | Note |
 | --- | --- | --- | --- | --- |
 | Federated muli-task learning under a mixture of distributions  | NeurIPS | [pdf](https://arxiv.org/pdf/2108.10252), [code](https://github.com/omarfoq/FedEM) | Distribution Mixture; Expectation-Maximization; FedEM |
-| Parameterized Knowledge Transfer for Personalized Federated Learning  | NeurIPS | [pdf](https://arxiv.org/pdf/2111.02862) | Knowledge Distillation |
+| Parameterized Knowledge Transfer for Personalized Federated Learning  | NeurIPS | [pdf](https://arxiv.org/pdf/2111.02862) | Knowledge Distillation |  transmit only soft-predictions; public dataset required
 | Personalized Federated Learning with Gaussian Processes | NeurIPS | [pdf](https://arxiv.org/pdf/2106.15482), [code](https://github.com/IdanAchituve/pFedGP) |  Gaussian process; Generalization bound |
 | Ditto: Fair and robust federated learning through personalization  | ICML | [pdf](https://arxiv.org/pdf/2012.04221), [code](https://github.com/litian96/ditto) | Threat model; Fairness; Regularizer |
 | Personalized Federated Learning using Hypernetworks  | ICML | [pdf](https://arxiv.org/pdf/2103.04628), [code](https://github.com/AvivSham/pFedHN) | Hypernetwork; Client Embedding |
-| Exploiting Shared Representations for Personalized Federated Learning  | ICML | [pdf](https://arxiv.org/pdf/2102.07078.pdf), [code](https://github.com/lgcollins/FedRep) | Partial model parameters |
+| Exploiting Shared Representations for Personalized Federated Learning  | ICML | [pdf](https://arxiv.org/pdf/2102.07078.pdf), [code](https://github.com/lgcollins/FedRep) | Partial model parameters | FedRep, shared body (feature extractor), personalized head (classifier)
 | Personalized Federated Learning with First Order Model Optimization  | ICLR | [pdf](https://arxiv.org/pdf/2012.08565), [code](https://github.com/NVlabs/FedFomo) | Model mixture |
 | FedBN: Federated Learning on Non-IID Features via Local Batch Normalization  | ICLR | [pdf](https://arxiv.org/pdf/2102.07623), [code](https://github.com/med-air/FedBN) | Partial model parameters |