ICLR

Adaptive Estimators Show Information Compression in Deep Neural Networks

Ivan Chelombiev, Conor Houghton, Cian O'Donnell

Comments Accepted as a poster presentation at ICLR 2019 and reviewed on OpenReview (available at https://openreview.net/forum?id=SkeZisA5t7). Pages: 11. Figures: 9

Multifactor Sequential Disentanglement via Structured Koopman Autoencoders

Nimrod Berman, Ilan Naiman, Omri Azencot

Comments Accepted to ICLR 2023, Notable-top 25% (Spotlight); The first two authors contributed equally

Conditional Antibody Design as 3D Equivariant Graph Translation

Xiangzhe Kong, Wenbing Huang, Yang Liu

Comments Accepted to ICLR 2023 as oral presentation. Outstanding paper honorable mentions

A View From Somewhere: Human-Centric Face Representations

Jerone T. A. Andrews, Przemyslaw Joniak, Alice Xiang

Comments Accepted to ICLR 2023. Code and data may be found at https://github.com/SonyAI/a_view_from_somewhere

Fairness-Aware Data Valuation for Supervised Learning

José Pombal, Pedro Saleiro, Mário A. T. Figueiredo, Pedro Bizarro

Comments ICLR 2023 Workshop Trustworthy ML

TempCLR: Temporal Alignment Representation with Contrastive Learning

Yuncong Yang, Jiawei Ma, Shiyuan Huang, Long Chen, Xudong Lin, Guangxing Han, Shih-Fu Chang

Comments ICLR 2023 Camera Ready. Code Link: https://github.com/yyuncong/TempCLR

An efficient encoder-decoder architecture with top-down attention for speech separation

Kai Li, Runxuan Yang, Xiaolin Hu

Comments Accepted by ICLR 2023; Code & Demos: https://cslikai.cn/project/TDANet/

Improving Small Molecule Generation using Mutual Information Machine

Danny Reidenbach, Micha Livne, Rajesh K. Ilango, Michelle Gill, Johnny Israeli

Comments Published at the MLDD workshop, ICLR 2023. version 2. 8 pages, 4 figures, 4 tables

Semi-Parametric Inducing Point Networks and Neural Processes

Richa Rastogi, Yair Schiff, Alon Hacohen, Zhaozhi Li, Ian Lee, Yuntian Deng, Mert R. Sabuncu, Volodymyr Kuleshov

Comments ICLR 2023 conference paper

Learning Iterative Neural Optimizers for Image Steganography

Xiangyu Chen, Varsha Kishore, Kilian Q Weinberger

Comments International Conference on Learning Representations (ICLR) 2023

Link Prediction with Non-Contrastive Learning

William Shiao, Zhichun Guo, Tong Zhao, Evangelos E. Papalexakis, Yozen Liu, Neil Shah

Comments ICLR 2023. 19 pages, 6 figures

Offline RL with No OOD Actions: In-Sample Learning via Implicit Value Regularization

Haoran Xu, Li Jiang, Jianxiong Li, Zhuoran Yang, Zhaoran Wang, Victor Wai Kin Chan, Xianyuan Zhan

Comments ICLR 2023 notable top 5%

One Transformer Can Understand Both 2D & 3D Molecular Data

Shengjie Luo, Tianlang Chen, Yixian Xu, Shuxin Zheng, Tie-Yan Liu, Liwei Wang, Di He

Comments 20 pages; ICLR 2023, Camera Ready Version; Code: https://github.com/lsj2408/Transformer-M

LPT: Long-tailed Prompt Tuning for Image Classification

Bowen Dong, Pan Zhou, Shuicheng Yan, Wangmeng Zuo

Comments ICLR 2023 (poster)

FINDE: Neural Differential Equations for Finding and Preserving Invariant Quantities

Takashi Matsubara, Takaharu Yaguchi

Comments 25 pages

Journal ref The Eleventh International Conference on Learning Representations (ICLR2023)

Probing optimisation in physics-informed neural networks

Nayara Fonseca, Veronica Guidetti, Will Trojak

Comments Accepted at the ICLR 2023 Workshop on Physics for Machine Learning

Sparse Distributed Memory is a Continual Learner

Trenton Bricken, Xander Davies, Deepak Singh, Dmitry Krotov, Gabriel Kreiman

Comments 9 Pages. ICLR Acceptance

Journal ref ICLR 2023

Task-Agnostic Graph Neural Network Evaluation via Adversarial Collaboration

Xiangyu Zhao, Hannes Stärk, Dominique Beaini, Yiren Zhao, Pietro Liò

Comments 11th International Conference on Learning Representations (ICLR 2023) Machine Learning for Drug Discovery (MLDD) Workshop. 17 pages, 6 figures, 4 tables

Autoregressive Conditional Neural Processes

Wessel P. Bruinsma, Stratis Markou, James Requiema, Andrew Y. K. Foong, Tom R. Andersson, Anna Vaughan, Anthony Buonomo, J. Scott Hosking, Richard E. Turner

Comments 57 pages; accepted to the 11th International Conference on Learning Representations (ICLR 2023)

An Equal-Size Hard EM Algorithm for Diverse Dialogue Generation

Yuqiao Wen, Yongchang Hao, Yanshuai Cao, Lili Mou

Comments Accepted by ICLR 2023

Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning

John Nguyen, Jianyu Wang, Kshitiz Malik, Maziar Sanjabi, Michael Rabbat

Comments Accepted at ICLR

Journal ref International Conference on Learning Representations 2023

Optimal Transport for Offline Imitation Learning

Yicheng Luo, Zhengyao Jiang, Samuel Cohen, Edward Grefenstette, Marc Peter Deisenroth

Comments Published in ICLR 2023

$k$NN Prompting: Beyond-Context Learning with Calibration-Free Nearest Neighbor Inference

Benfeng Xu, Quan Wang, Zhendong Mao, Yajuan Lyu, Qiaoqiao She, Yongdong Zhang

Comments ICLR 2023. Code is available at https://github.com/BenfengXu/KNNPrompting

Journal ref ICLR 2023

Temperature Schedules for Self-Supervised Contrastive Methods on Long-Tail Data

Anna Kukleva, Moritz Böhle, Bernt Schiele, Hilde Kuehne, Christian Rupprecht

Comments ICLR 2023

When and why vision-language models behave like bags-of-words, and what to do about it?

Mert Yuksekgonul, Federico Bianchi, Pratyusha Kalluri, Dan Jurafsky, James Zou

Comments ICLR 2023 Oral (notable-top-5%)

Query Efficient Decision Based Sparse Attacks Against Black-Box Deep Learning Models

Viet Quoc Vo, Ehsan Abbasnejad, Damith C. Ranasinghe

Comments Published as a conference paper at the International Conference on Learning Representations (ICLR 2022). Code is available at https://sparseevoattack.github.io/

DeBERTaV3: Improving DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled Embedding Sharing

Pengcheng He, Jianfeng Gao, Weizhu Chen

Comments 16 pages, 10 tables, 2 Figures. The DeBERTaV3 model significantly improves performance of the downstream NLU tasks over models with a similar structure, e.g. DeBERTaV3 large achieves 91.37% average GLUE score which is 1.37% over DeBERTa large. XSmall has only 22M backbone parameters, but significantly outperforms RoBERTa/XLNet-base. Paper is published as a conference paper at ICLR 2023

A Closer Look at Model Adaptation using Feature Distortion and Simplicity Bias

Puja Trivedi, Danai Koutra, Jayaraman J. Thiagarajan

Comments Accepted to ICLR 2023 as notable-25% (spotlight)

Symmetries, flat minima, and the conserved quantities of gradient flow

Bo Zhao, Iordan Ganev, Robin Walters, Rose Yu, Nima Dehmamy

Comments To appear at ICLR 2023

Learning Low Dimensional State Spaces with Overparameterized Recurrent Neural Nets

Edo Cohen-Karlik, Itamar Menuhin-Gruman, Raja Giryes, Nadav Cohen, Amir Globerson

Comments Accepted to ICLR 2023, 9 pages, 2 figures plus supplementary