ICML2022大会论文列表（1）

ICML 是 International Conference on Machine Learning的缩写，即国际机器学习大会。ICML如今已发展为由国际机器学习学会（IMLS）主办的年度机器学习国际顶级会议。

ICML2022共收到5630 篇投稿，其中，1117 篇被接收为short oral，118篇被接收为long oral。接收率为21.94%，与以往几年基本持平。本届大会共评选出15 篇杰出论文奖和 1 项时间检验奖。其中，复旦大学、上海交通大学、厦门大学、莱斯大学等多个华人团队的工作被评位杰出论文奖。ICML 2012 的一篇论文《Poisoning Attacks against Support Vector Machines》获得了时间检验奖。

获奖论文信息详见：https://icml.cc/virtual/2022/awards_detail

01. PAC-Bayesian Bounds on Rate-Efficient Classifiers

作者：Alhabib Abbas,Yiannis Andreopoulos

原文地址：

https://proceedings.mlr.press/v162/abbas22a/abbas22a.pdf

02. Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learning

作者：Momin Abbas,Quan Xiao, Lisha Chen, Pin-Yu Chen, Tianyi Chen

原文地址：

https://proceedings.mlr.press/v162/abbas22b/abbas22b.pdf

Github:

https://github.com/mominabbass/sharp-maml

03. An Initial Alignment between Neural Network and Target is Needed for Gradient Descent to Learn

作者：Emmanuel Abbe,Elisabetta Cornacchia, Jan Hazla, Christopher Marquis

原文地址：

https://proceedings.mlr.press/v162/abbe22a/abbe22a.pdf

04. Active Sampling for Min-Max Fairness

作者：Jacob D Abernethy,Pranjal Awasthi, Matthäus Kleindessner, Jamie Morgenstern, Chris Russell, Jie Zhang

原文地址：

https://proceedings.mlr.press/v162/abernethy22a/abernethy22a.pdf

Github:

https://github.com/amazon-research/active-sampling-for-minmax-fairness

05. Meaningfully debugging model mistakes using conceptual counterfactual explanations

作者：Abubakar Abid,Mert Yuksekgonul, James Zou

原文地址：

https://proceedings.mlr.press/v162/abid22a/abid22a.pdf

Github:

https://github.com/mertyg/debug-mistakes-cce

06. Batched Dueling Bandits

作者：Arpit Agarwal,Rohan Ghuge, Viswanath Nagarajan

原文地址：

https://proceedings.mlr.press/v162/agarwal22a/agarwal22a.pdf

07. Hierarchical Shrinkage: Improving the accuracy and interpretability of tree-based models

作者：Abhineet Agarwal,Yan Shuo Tan, Omer Ronen, Chandan Singh, Bin Yu

原文地址：

https://proceedings.mlr.press/v162/agarwal22b/agarwal22b.pdf

Github：

https://github.com/yu-group/imodels-experiments

https://github.com/csinva/imodels

08. Deep equilibrium networks are sensitive to initialization statistics

作者：Atish Agarwala,Samuel S Schoenholz

原文地址：

https://proceedings.mlr.press/v162/agarwala22a/agarwala22a.pdf

09. Learning of Cluster-based Feature Importance for Electronic Health Record Time-series

作者：Henrique Aguiar,Mauro Santos, Peter Watkinson, Tingting Zhu

原文地址：

https://proceedings.mlr.press/v162/aguiar22a/aguiar22a.pdf

10. On the Convergence of the Shapley Value in Parametric Bayesian Learning Games

作者：Lucas Agussurja,Xinyi Xu, Bryan Kian Hsiang Low

原文地址：

https://proceedings.mlr.press/v162/agussurja22a/agussurja22a.pdf

Github：

https://github.com/XinyiYS/Parametric-Bayesian-Learning-Games

11. Individual Preference Stability for Clustering

作者：Saba Ahmadi,Pranjal Awasthi, Samir Khuller, Matthäus Kleindessner, Jamie Morgenstern, Pattara Sukprasert, Ali Vakilian

原文地址：

https://proceedings.mlr.press/v162/ahmadi22a/ahmadi22a.pdf

Github：

https://github.com/amazon-research/ip-stability-for-clustering

12. Understanding the unstable convergence of gradient descent

作者：Kwangjun Ahn,Jingzhao Zhang, Suvrit Sra

原文地址：

https://proceedings.mlr.press/v162/ahn22a/ahn22a.pdf

13. Minimum Cost Intervention Design for Causal Effect Identification

作者：Sina Akbari,Jalal Etesami, Negar Kiyavash

原文地址：

https://proceedings.mlr.press/v162/akbari22a/akbari22a.pdf

其他材料：

https://media.icml.cc/Conferences/ICML2022/supplementary/akbari22a-supp.zip

GitHub：

https://github.com/sinaakbarii/min_cost_intervention

14. How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Models

作者：Ahmed Alaa,Boris Van Breugel, Evgeny S. Saveliev, Mihaela van der Schaar

原文地址：

https://proceedings.mlr.press/v162/alaa22a/alaa22a.pdf

GitHub：

https://github.com/vanderschaarlab/evaluating-generative-models

15. A Natural Actor-Critic Framework for Zero-Sum Markov Games

作者：Ahmet Alacaoglu,Luca Viano, Niao He, Volkan Cevher

原文地址：

https://proceedings.mlr.press/v162/alacaoglu22a/alacaoglu22a.pdf

16. Deploying Convolutional Networks on Untrusted Platforms Using 2D Holographic Reduced Representations

作者：Mohammad Mahmudul Alam,Edward Raff, Tim Oates, James Holt

原文地址：

https://proceedings.mlr.press/v162/alam22a/alam22a.pdf

GitHub：

https://github.com/neuromorphiccomputationresearchprogram/connectionist-symbolic-pseudo-secrets

17. Optimistic Linear Support and Successor Features as a Basis for Optimal Policy Transfer

作者：Lucas Nunes Alegre,Ana Bazzan, Bruno C. Da Silva

原文地址：

https://proceedings.mlr.press/v162/alegre22a/alegre22a.pdf

GitHub：

https://github.com/lucasalegre/sfols

18. Structured Stochastic Gradient MCMC

作者：Antonios Alexos,Alex J Boyd, Stephan Mandt

原文地址：

https://proceedings.mlr.press/v162/alexos22a/alexos22a.pdf

github：

https://github.com/ajboyd2/pytorch_lvi

19. XAI for Transformers: Better Explanations through Conservative Propagation

作者：Ameen Ali,Thomas Schnake, Oliver Eberle, Grégoire Montavon, Klaus-Robert Müller, Lior Wolf

原文地址：

https://proceedings.mlr.press/v162/ali22a/ali22a.pdf

其他材料：

https://media.icml.cc/Conferences/ICML2022/supplementary/ali22a-supp.zip

Github：

https://github.com/ameenali/xai_transformers

20. RUMs from Head-to-Head Contests

作者：Matteo Almanza,Flavio Chierichetti, Ravi Kumar, Alessandro Panconesi, Andrew Tomkins

原文地址：

https://proceedings.mlr.press/v162/almanza22a/almanza22a.pdf

其他材料：

https://media.icml.cc/Conferences/ICML2022/supplementary/almanza22a-supp.zip

21. Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval

作者：Uri Alon,Frank Xu, Junxian He, Sudipta Sengupta, Dan Roth, Graham Neubig

原文地址：

https://proceedings.mlr.press/v162/alon22a/alon22a.pdf

GitHub：

https://github.com/neulab/retomaton

22. Minimax Classification under Concept Drift with Multidimensional Adaptation and Performance Guarantees

作者：Verónica Álvarez,Santiago Mazuelas, Jose A Lozano

原文地址：

https://proceedings.mlr.press/v162/alvarez22a/alvarez22a.pdf

其他材料：

https://media.icml.cc/Conferences/ICML2022/supplementary/alvarez22a-supp.zip

GitHub：

https://github.com/machinelearningbcam/amrc-for-concept-drift-icml-2022

23. Scalable First-Order Bayesian Optimization via Structured Automatic Differentiation

作者：Sebastian E Ament,Carla P Gomes

原文地址：

https://proceedings.mlr.press/v162/ament22a/ament22a.pdf

GitHub：

https://github.com/sebastianament/covariancefunctions.jl

24. Public Data-Assisted Mirror Descent for Private Model Training

作者：Ehsan Amid,Arun Ganesh, Rajiv Mathews, Swaroop Ramaswamy, Shuang Song, Thomas Steinke, Thomas Steinke, Vinith M Suriyakumar, Om Thakkar, Abhradeep Thakurta

原文地址：

https://proceedings.mlr.press/v162/amid22a/amid22a.pdf

25. On Last-Iterate Convergence Beyond Zero-Sum Games

作者：Ioannis Anagnostides,Ioannis Panageas, Gabriele Farina, Tuomas Sandholm

原文地址：

https://proceedings.mlr.press/v162/anagnostides22a/anagnostides22a.pdf

26. Online Algorithms with Multiple Predictions

作者：Keerti Anand,Rong Ge, Amit Kumar, Debmalya Panigrahi

原文地址：

https://proceedings.mlr.press/v162/anand22a/anand22a.pdf

27. Learning to Hash Robustly, Guaranteed

作者：Alexandr Andoni,Daniel Beaglehole

原文地址：

https://proceedings.mlr.press/v162/andoni22a/andoni22a.pdf

28. Set Based Stochastic Subsampling

作者：Bruno Andreis,Seanie Lee, A. Tuan Nguyen, Juho Lee, Eunho Yang, Sung Ju Hwang

原文地址：

https://proceedings.mlr.press/v162/andreis22a/andreis22a.pdf

29. Towards Understanding Sharpness-Aware Minimization

作者：Maksym Andriushchenko,Nicolas Flammarion

原文地址：

https://proceedings.mlr.press/v162/andriushchenko22a/andriushchenko22a.pdf

Github：

https://github.com/tml-epfl/understanding-sam

30. Fair and Fast k-Center Clustering for Data Summarization

作者：Haris Angelidakis,Adam Kurpisz, Leon Sering, Rico Zenklusen

原文地址：

https://proceedings.mlr.press/v162/angelidakis22a/angelidakis22a.pdf

其他材料：

https://media.icml.cc/Conferences/ICML2022/supplementary/angelidakis22a-supp.zip