Learning the Arrow of Time for Human Activity Recognition: A Study on 700,000 Person-days of Activity Tracker Data - [Poster Session II] [Zoom Link] Hang Yuan (University of Oxford)*; Shing Chan (University of Oxford); Aiden Doherty (University of Oxford)

Continual-wav2vec2: an Application of Continual Learning for Self-Supervised Automatic Speech Recognition - [Poster Session I] [Zoom Link] Samuel Kessler (Oxford University)*; Bethan J Thomas (Huawei R&D UK); Salah Karout (Huawei)

Anomalous Region Detection in Time Series with Local Neural Transformations -[Poster Session I] [Zoom Link] Tim Schneider (Bosch Center for Artificial Intelligence)*; Chen Qiu (Bosch Center for Artificial Intelligence); Maja Rudolph (Bosch Center for Artificial Intelligence)

Abstract Reasoning via Logic-guided Generation - [Poster Session II] [Zoom Link] Sihyun Yu (KAIST)*; Sangwoo Mo (KAIST); Sungsoo Ahn (KAIST); Jinwoo Shin (KAIST)

Reinforcement Learning as One Big Sequence Modeling Problem - [Poster Session I] [Zoom Link] Michael Janner (UC Berkeley)*; Qiyang Li (University of California, Berkeley); Sergey Levine (UC Berkeley)

Contrastive Self-supervised BERT for Vision and Language Pre-training - [Poster Session II] [Zoom Link] Shentong Mo (Carnegie Mellon University)*; Jingfei Xia (Carnegie Mellon University ); Ihor Markevych (Carnegie Mellon University)

Unsupervised Disentanglement without Autoencoding: Pitfalls and Future Directions - [Poster Session I] [Zoom Link] Andrea Burns (Boston University)*; Aaron Sarna (Google); Dilip Krishnan (Google); Aaron Maschinot (Google Research)

Graph Contrastive Pre-training for Effective Theorem Reasoning - [Poster Session II] [Zoom Link] Zhaoyu Li (SJTU)*; Xujie Si (McGill University); Binghong Chen (Georgia Tech)

Transforming Self-Supervision in Test Time for Personalizing Human Pose Estimation - [Poster Session I] [Zoom Link] Miao Hao (University of California San Diego); Yizhuo Li (Shanghai Jiao Tong University)*; Zonglin Di (University of California, San Diego); Nitesh Gundavarapu (University of California San Diego); Xiaolong Wang (UCSD)

TTT++: When Does Self-supervised Test-time Training Fail or Thrive? - [Poster Session I] [Zoom Link] Yuejiang Liu (EPFL)*; Parth Kothari (Ecole Polytechnique Fédérale de Lausanne); Bastien Van Delft (Ecole Polytechnique Fédérale de Lausanne); Baptiste Bellot-Gurlet (Ecole Polytechnique Fédérale de Lausanne); Taylor Mordan (EPFL); Alexandre Alahi (EPFL)

NetVec: A Hierarchical Hypergraph Embedding System - [Poster Session I] [Zoom ID: 304 484 9402 PWD: 863500] sepideh maleki (The University of Texas at Austin)*; Donya Saless (University of Tehran); Dennis Wall (Stanford University); Keshav Pingali (The University of Texas at Austin)

Self-Supervised Neural Architecture Search for Imbalanced Datasets - [Poster Session I] [Zoom Link] Aleksandr Timofeev (EPFL)*; Grigorios Chrysos (École Polytechnique Fédérale de Lausanne); Volkan Cevher (EPFL)

SIMONe: View-Invariant, Temporally-Abstracted Object Representations via Unsupervised Video Decomposition - [Poster Session I] [Zoom Link] Rishabh Kabra (DeepMind)*; Daniel Zoran (DeepMind); Goker Erdogan (DeepMind); Loic Matthey (DeepMind); Antonia Creswell (DeepMind); Matthew Botvinick (google); Alexander Lerchner (DeepMind); Chris Burgess (Wayve)

Disentangling ODE parameters from dynamics in VAEs - [Poster Session I] [Zoom Link] Stathi Fotiadis (Imperial College London)*; Anil Anthony Bharath ((Imperial College of London,UK)); Chris Cantwell (Imperial College London); Mario A Lino (Imperial College London)

The Impact of Negative Sampling on Contrastive Structured World Models - [Poster Session I] [Zoom Link] Ondrej Biza (Northeastern University)*; Elise van der Pol (University of Amsterdam); Thomas Kipf (Google Brain)

Discovering and Achieving Goals with World Models - [Poster Session I] [Zoom Link] Russell Mendonca (Carnegie Mellon University); Oleh Rybkin (University of Pennsylvania)*; Kostas Daniilidis (University of Pennsylvania); Danijar Hafner (Google); Deepak Pathak (Carnegie Mellon University)

Constellation: Learning relational abstractions over objects for compositional imagination - [Poster Session I] [Zoom Link] James Whittington (University of Oxford)*; Rishabh Kabra (DeepMind); Loic Matthey (DeepMind); Chris Burgess (Wayve); Alexander Lerchner (DeepMind)

Beyond Learning Features: Training a Fully-functional Classifier with ZERO Instance-level Labels -[Poster Session I] [Zoom Link] deepak babu sam (Indian Institute of Science)*; Abhinav Agarwalla (Carnegie Mellon University); Venkatesh Babu RADHAKRISHNAN (Indian Institute of Science)

Self-Supervised Learning with Data Augmentations Provably Isolates Content from Style - [Poster Session I] [Zoom Link] Julius von Kügelgen (MPI for Intelligent Systems, Tübingen & University of Cambridge)*; Yash Sharma (University of Tuebingen / CIN); Luigi Gresele (MPI for Intelligent Systems, Tübingen); Wieland Brendel (University of Tübingen); Bernhard Schölkopf (MPI for Intelligent Systems, Tübingen); Michel Besserve (MPI for Intelligent Systems, Tübingen); Francesco Locatello (Amazon)

Scaling up Logical Query Embeddings on Knowledge Graphs - [Poster Session I] [Zoom Link] Hongyu Ren (Stanford University)*; Hanjun Dai (Google Brain); Bo Dai (Google Brain); Xinyun Chen (UC Berkeley); Denny Zhou (Google Brain); Jure Leskovec (Stanford University); Dale Schuurmans (Google / University of Alberta)

Batch Curation for Unsupervised Contrastive Representation Learning - [Poster Session I] [Zoom Link] Michael Welle (KTH Royal Institute of Technology)*; Petra Poklukar (KTH Royal Institute of Technology); Danica Kragic (KTH Royal Institute of Technology)

Don’t Sweep your Learning Rate under the Rug: A Closer Look at Cross-modal Transfer of Pretrained Transformers - [Poster Session I] [Zoom Link] Danielle Rothermel (Facebook AI Research)*; Margaret Li (University of Washington); Tim Rocktäschel (Facebook AI Research & University College London); Jakob Foerster (Facebook AI Research)

Physical Reasoning Using Dynamics-Aware Models - [Poster Session I] [Zoom Link]Passcode: 554932 Eltayeb K. E. Ahmed (Facebook)*; Anton Bakhtin (Facebook AI Research); Laurens van der Maaten (Facebook); Rohit Girdhar (Facebook AI Research)

Towards Fine-grained Visual Representations by Combining Contrastive Learning with Image Reconstruction and Attention-weighted Pooling - [Poster Session I] [Zoom Link]Jonas Dippel (Bayer AG)*; Steffen Vogler (Bayer AG); Johannes Höhne (Bayer AG)

Unsupervised Object-Based Transition Models For Embodied Agents in 3D Partially Observable Environments - [Poster Session I] [Zoom Link]Passcode: OAT Antonia Creswell (DeepMind)*; Rishabh Kabra (DeepMind); Chris Burgess (DeepMind); Murray P Shanahan (DeepMind Technologies Ltd)

Evaluation Criteria for Deep Clustering Algorithms - [Poster Session I] [Zoom Link]Passcode: 815100 Jayanth R Regatti (The Ohio State University)*; Aniket Anand Deshmukh (Microsoft); Eren Manavoglu (Microsoft); Urun Dogan (Microsoft Research)

Improving Transferability of Representations via Augmentation-Aware Self-Supervision - [Poster Session II] [Zoom Link] Hankook Lee (KAIST)*; Kibok Lee (University of Michigan); Kimin Lee (UC Berkeley); Honglak Lee (LG AI Research / University of Michigan); Jinwoo Shin (KAIST)

Learning Self-supervised 3D Features from Single-view Depth Scans - [Poster Session I] [Zoom Link]Zaiwei Zhang (University of Texas at Austin)*; Rohit Girdhar (Facebook AI Research); Armand Joulin (Facebook AI Research); Ishan Misra (Facebook AI Research)

Contrastive Learning of Image Representations with Cross-Video Cycle-Consistency - [Poster Session I] [Zoom Link] Haiping Wu (McGill University)*; Xiaolong Wang (UCSD)

Intermediate Layers Matter in Momentum Contrastive Self Supervised Learning - [Poster Session II] [Zoom Link] Aakash Kaku (NYU Center for Data Science)*; Sahana Upadhya (New York University); Narges Razavian (New York University School of Medicine)

Rethinking Self-supervised Correspondence Learning: A Video Frame-level Similarity Perspective - [Poster Session II] [Zoom Link]Jiarui Xu (University of California San Diego)*; Xiaolong Wang (UCSD)

KB-VLP: Knowledge Based Vision and Language Pretraining - [Poster Session I] [Zoom Link] Kezhen Chen (Northwestern University)*; Qiuyuan Huang (Microsoft Research, Redmond); Yonatan Bisk (Carnegie Mellon University); Daniel McDuff (Microsoft Research); Jianfeng Gao (Microsoft Research)

Aux-AIRL: End-to-End Self-Supervised Reward Learning for Extrapolating beyond Suboptimal Demonstrations - [Poster Session II] [Zoom Link] Yuchen Cui (University of Texas at Austin)*; Bo Liu (University of Texas, Austin); Akanksha Saran (University of Texas at Austin); Stephen Giguere (University of Texas at Austin); Scott Niekum (UT Austin); Peter Stone (University of Texas at Austin and Sony AI)

Self-Supervised Contrastive Feature Learning for Whole Slide Image Classification - [Poster Session I] [Zoom Link] Bin Li (University of Wisconsin-Madison)*; Yin Li (University of Wisconsin-Madison); Kevin Eliceiri (University of Wisconsin - Madison)

Scalable pragmatic communication via self-supervision - [Poster Session I] [Zoom Link] Jennifer Hu (Massachusetts Institute of Technology)*; Roger Levy (Massachusetts Institute of Technology); Noga Zaslavsky (MIT)

Reasoning-Modulated Representations - [Poster Session I] [Zoom Link]Petar Veličković (DeepMind)*; Matko Bošnjak (DeepMind); Thomas Kipf (Google Brain); Alexander Lerchner (DeepMind); Raia Hadsell (Deepmind); Razvan Pascanu (Google Deepmind); Charles Blundell (DeepMind)

Can contrastive learning avoid shortcut solutions? - [Poster Session I] [Zoom Link]Joshua D Robinson (MIT)*; Li Sun (University of Pittsburgh); Ke Yu (University of Pittsburgh); Kayhan Batmanghelich (University of Pittsburgh); Stefanie Jegelka (MIT); Suvrit Sra (Massachusetts Institute of Technology, USA)

BMWReg: Brownian-diffusive, Multiview, Whitening Regulararizations forSelf-supervised Learning - [Poster Session I] [Zoom Link] Suhong Moon (UC Berkeley); Domas Buracas (UC Berkeley); Jinkyu Kim (Korea University); John F Canny (UC Berkeley)*

Contrastive Spatial Reasoning on Multi-View Line Drawings - [Poster Session I] [Zoom Link] Siyuan Xiang (New York University); Anbang Yang (New York University); Yanfei Xue (New York University); Yaoqing Yang (UC Berkeley); Chen Feng (New York University)*

Hough Contrastive Learning: Hough Voting to Find Better Positive Pairs in Dense Self-supervised Representation Learning - [Poster Session II] [Zoom Link] Yunsung Lee (Korea University)*; Teakgyu Hong (NAVER Corp.); Han-Cheol Cho (Naver); Junbum Cha (Clova AI Research, NAVER Corp.); Seungryong Kim (Korea University)

Group Supervised Learning: Extending Self-Supervised Learning to Multi-Device Settings - [Poster Session I] [Zoom Link] Yash Jain (Indian Institute of Technology Bombay); Chi Ian Tang (University of Cambridge); Chulhong Min (Nokia Bell Labs); Fahim Kawsar (Nokia Bell Labs); Akhil Mathur (Nokia Bell Labs)*

Fuzzy Logic based Logical Query Answering on Knowledge Graphs - [Poster Session II] [Zoom Link] Xuelu Chen (UCLA)*; Ziniu Hu (UCLA); Yizhou Sun (UCLA)

A Probabilistic Interpretation of Transformers - [Poster Session II] [Zoom Link] Alexander J Shim (ML Collective)