Accepted papers


  • 3D Group-Equivariant Neural Networks for Octahedral and Square Prism Symmetry Groups.  (Oral presentation)
    Marysia Winkels*, Aidence/University of Amsterdam; Taco Cohen, University of Amsterdam
  • Learning representations that account for data symmetries.   (Oral presentation)
    Fabio Anselmi*, "MIT, IIT"; Georgios Evangelopoulos, X, Alphabet; Tomaso Poggio, MIT; Lorenzo Rosasco 
  • Scale equivariance in CNNs with vector fields.   (Oral presentation)
    Diego Marcos*, University of Wageningen; Benjamin Kellenberger, Wageningen University and Research; Sylvain Lobry, Wageningen University and Research; Devis Tuia, Wageningen University and Research 
  • Universal approximations of invariant maps by neural networks.   (Oral presentation)
    Dmitry Yarotsky*, Skolkovo Institute of Science and Technology 
  • A Kernel Theory of Modern Data Augmentation.   (Oral presentation)
    Tri Dao*, Stanford University; Albert Gu, Stanford University; Alexander Ratner, Stanford University; Virginia Smith, Stanford; Chris De Sa, Cornell; Christopher Re, Stanford University 
  • Planning with Arithmetic and Geometric Attributes.   (Oral presentation)
    David Folque, NYU; Sainbayar Sukhbaatar, NYU; Arthur Szlam, Facebook; Joan Bruna*, Courant Institute of Mathematical Sciences, NYU, USA 
  • Sample Efficient Semantic Segmentation using Rotation Equivariant Convolutional Networks.   
    Jasper Linmans*, University of Amsterdam; Jim Winkens, University of Amsterdam; Bastiaan Veeling, University of Amsterdam; Taco Cohen, University of Amsterdam; Max Welling, University of Amsterdam
  • Robustness of Rotation-Equivariant Networks to Adversarial Perturbations.   
    Beranger Dumont, Rakuten; Simona Maggio, Rakuten; Pablo Montalvo*, Rakuten
  • Prediction Propagation for Domain Adaptation in NLP.   
    Bjarke Felbo*, MIT; Michiel Bakker, MIT; Abhimanyu Dubey, Massachusetts Institute of Technology; Sadhika Malladi, Massachusetts Institute of Technology; Alex `Sandy' Pentland, MIT; Iyad Rahwan, MIT
  • Gradient Regularization Improves Accuracy of Discriminative Models. 
    Dániel Varga*, Alfréd Rényi Institute of Mathematics; Adrián Csiszárik, Alfréd Rényi Institute of Mathematics; Zsolt Zombori, Rényi Institute of Mathematics
  • Semantic Segmentation with Scarce Data. 
    Isay Katsman*, Cornell University; Rohun Tripathi, Cornell; Andreas Veit, Cornell University; Serge Belongie, Cornell University
  • Matrix Co-completion for Multi-label Classification with Missing Features and Labels.   
    Miao Xu*, RIKEN; Gang Niu, RIKEN; Bo Han, University of Technology Sydney; Ivor Tsang, University of Technology Sydney; Zhi-Hua Zhou, Nanjing university; Masashi Sugiyama, RIKEN/The University of Tokyo
  • Understanding Adversarial Robustness of Symmetric Networks. 
    Sandesh Kamath*, Chennai Mathematical Institute; Amit Deshpande, Microsoft Research
  • Learning Invariances using the Marginal Likelihood. 
    Mark van der Wilk*, PROWLER.io; Matthias Bauer, ; ST John, PROWLER.io; James Hensman, PROWLER.io
  • Comparison of Data Efficiency in Dynamic Routing for Capsule Networks.   Kenny Schlegel, TU Chemnitz; Peer Neubert*, TU Chemnitz; Peter Protzel, TU Chemnitz
  • Zero Shot Adversarial Data Programming:Using GANs to Relax the Bottleneck of Curated Labeled Data.  
    Arghya Pal*,  Indian Institute of Technology Hyderabad; Vineeth N Balasubramanian, IIT Hyderabad
  • Learning Compressed Transforms with Low Displacement Rank.  
    Anna Thomas*, Stanford; Albert Gu, Stanford University; Tri Dao, Stanford University; Atri Rudra, University at Buffalo, SUNY; Christopher Re, Stanford University
  • Zero-Shot Image Generation by Distilling Concepts from Multiple Captions.   
    Joseph K J*, Indian Institute of Technology, Hyderabad; Arghya Pal,  Indian Institute of Technology Hyderabad; Vineeth N Balasubramanian, IIT Hyderabad
  • CoNet: Collaborative Cross Networks for Cross-Domain Recommendation.
    Guangneng Hu*, HKUST