ICML

NeuralStagger: Accelerating Physics-constrained Neural PDE Solver with Spatial-temporal Decomposition

Xinquan Huang, Wenlei Shi, Qi Meng, Yue Wang, Xiaotian Gao, Jia Zhang, Tie-Yan Liu

Comments ICML 2023 accepted

Multi-Fidelity Covariance Estimation in the Log-Euclidean Geometry

Aimee Maurais, Terrence Alsup, Benjamin Peherstorfer, Youssef Marzouk

Comments To appear at the International Conference on Machine Learning (ICML) 2023

Variational sparse inverse Cholesky approximation for latent Gaussian processes via double Kullback-Leibler minimization

Jian Cao, Myeongjong Kang, Felix Jimenez, Huiyan Sang, Florian Schafer, Matthias Katzfuss

Comments Accepted at the 2023 International Conference on Machine Learning (ICML). 18 pages with references and appendices, 14 figures

Estimating Causal Effects using a Multi-task Deep Ensemble

Ziyang Jiang, Zhuoran Hou, Yiling Liu, Yiman Ren, Keyu Li, David Carlson

Comments 18 pages, 7 figures, 3 tables, published at the 40th International Conference on Machine Learning (ICML 2023)

Understanding the Impact of Adversarial Robustness on Accuracy Disparity

Yuzheng Hu, Fan Wu, Hongyang Zhang, Han Zhao

Comments Accepted at ICML 2023

PAC-Bayesian Offline Contextual Bandits With Guarantees

Otmane Sakhi, Pierre Alquier, Nicolas Chopin

Comments Accepted to ICML 2023

VIMA: General Robot Manipulation with Multimodal Prompts

Yunfan Jiang, Agrim Gupta, Zichen Zhang, Guanzhi Wang, Yongqiang Dou, Yanjun Chen, Li Fei-Fei, Anima Anandkumar, Yuke Zhu, Linxi Fan

Comments ICML 2023 Camera-ready version. Project website: https://vimalabs.github.io/

Learning to acquire novel cognitive tasks with evolution, plasticity and meta-meta-learning

Thomas Miconi

Journal ref 40th International Conference on Machine Learning (ICML 2023)

A Modern Take on the Bias-Variance Tradeoff in Neural Networks

Brady Neal, Sarthak Mittal, Aristide Baratin, Vinayak Tantia, Matthew Scicluna, Simon Lacoste-Julien, Ioannis Mitliagkas

Journal ref ICML 2019 Workshop on Identifying and Understanding Deep Learning Phenomena

Future-conditioned Unsupervised Pretraining for Decision Transformer

Zhihui Xie, Zichuan Lin, Deheng Ye, Qiang Fu, Wei Yang, Shuai Li

Comments 17 pages, 9 figures, ICML 2023

Emergent Agentic Transformer from Chain of Hindsight Experience

Hao Liu, Pieter Abbeel

Comments International Conference on Machine Learning (ICML) 2023

Coarse-to-Fine: a Hierarchical Diffusion Model for Molecule Generation in 3D

Bo Qiang, Yuxuan Song, Minkai Xu, Jingjing Gong, Bowen Gao, Hao Zhou, Weiying Ma, Yanyan Lan

Comments ICML 2023 poster

I$^2$SB: Image-to-Image Schrödinger Bridge

Guan-Horng Liu, Arash Vahdat, De-An Huang, Evangelos A. Theodorou, Weili Nie, Anima Anandkumar

Comments ICML camera ready (high-resolution figures)

Implicit Regularization Leads to Benign Overfitting for Sparse Linear Regression

Mo Zhou, Rong Ge

Comments ICML 2023 camera ready version

Maximal Initial Learning Rates in Deep ReLU Networks

Gaurav Iyer, Boris Hanin, David Rolnick

Comments International Conference on Machine Learning (ICML) 2023

Neural networks trained with SGD learn distributions of increasing complexity

Maria Refinetti, Alessandro Ingrosso, Sebastian Goldt

Comments Source code available at https://github.com/sgoldt/dist_inc_comp

Journal ref ICML 2023

Everyone's Preference Changes Differently: Weighted Multi-Interest Retrieval Model

Hui Shi, Yupeng Gu, Yitong Zhou, Bo Zhao, Sicun Gao, Jishen Zhao

Comments Accepted by ICML 23'

Fascinating Supervisory Signals and Where to Find Them: Deep Anomaly Detection with Scale Learning

Hongzuo Xu, Yijie Wang, Juhui Wei, Songlei Jian, Yizhou Li, Ning Liu

Comments Accepted by ICML 2023

Detecting Adversarial Data by Probing Multiple Perturbations Using Expected Perturbation Score

Shuhai Zhang, Feng Liu, Jiahao Yang, Yifan Yang, Changsheng Li, Bo Han, Mingkui Tan

Comments Accepted at ICML 2023

Minimizing Trajectory Curvature of ODE-based Generative Models

Sangyun Lee, Beomsu Kim, Jong Chul Ye

Comments ICML 2023

SOM-CPC: Unsupervised Contrastive Learning with Self-Organizing Maps for Structured Representations of High-Rate Time Series

Iris A. M. Huijben, Arthur A. Nijdam, Sebastiaan Overeem, Merel M. van Gilst, Ruud J. G. van Sloun

Journal ref International Conference on Machine Learning 2023

Sample and Predict Your Latent: Modality-free Sequential Disentanglement via Contrastive Estimation

Ilan Naiman, Nimrod Berman, Omri Azencot

Comments Accepted to ICML 2023; The first two authors contributed equally

On the Impact of Knowledge Distillation for Model Interpretability

Hyeongrok Han, Siwon Kim, Hyun-Soo Choi, Sungroh Yoon

Comments International Conference on Machine Learning (ICML) 2023

Chameleon: Adapting to Peer Images for Planting Durable Backdoors in Federated Learning

Yanbo Dai, Songze Li

Comments This paper was accepted to ICML 2023

Near-Minimax-Optimal Risk-Sensitive Reinforcement Learning with CVaR

Kaiwen Wang, Nathan Kallus, Wen Sun

Comments Accepted at ICML 2023

A theory of continuous generative flow networks

Salem Lahlou, Tristan Deleu, Pablo Lemos, Dinghuai Zhang, Alexandra Volokhova, Alex Hernández-García, Léna Néhale Ezzine, Yoshua Bengio, Nikolay Malkin

Comments ICML 2023; 32 pages; code: https://github.com/saleml/continuous-gfn

Mastering the Unsupervised Reinforcement Learning Benchmark from Pixels

Sai Rajeswar, Pietro Mazzaglia, Tim Verbelen, Alexandre Piché, Bart Dhoedt, Aaron Courville, Alexandre Lacoste

Comments Accepted at ICML 2023 (oral)

A theory of representation learning gives a deep generalisation of kernel methods

Adam X. Yang, Maxime Robeyns, Edward Milsom, Ben Anson, Nandi Schoots, Laurence Aitchison

Comments Published in ICML 2023

Masked Bayesian Neural Networks : Theoretical Guarantee and its Posterior Inference

Insung Kong, Dongyoon Yang, Jongjin Lee, Ilsang Ohn, Gyuseung Baek, Yongdai Kim

Comments 30 pages, ICML 2023 proceedings. arXiv admin note: substantial text overlap with arXiv:2206.00853

Towards credible visual model interpretation with path attribution

Naveed Akhtar, Muhammad A. A. K. Jalwana

Comments ICML'23 paper (text improved for CV community)