Causality Causality Workbench                                                             Challenges in Machine Learning Causality

Unsupervised and Transfer Learning Challenge

  • The unsupervised and transfer learning challenge is part of the IJCNN 2011 conference competition program and the Pascal2 challenges.

    INNS IEEE CIS Pascal2
  • We are very grateful to our other sponsors who made this project possible:

    DL NRL Orange Clopinet MisterP Services ETH Unipen HDC NEC Kaggle

  • Data donors:
    This project would not have been possible without generous donations of data. We are very grateful to the data donors:
    Handwriting recognition (AVICENNA) -- Reza Farrahi Moghaddam, Mathias Adankon, Kostyantyn Filonenko, Robert Wisnovsky, and Mohamed Chériet (Ecole de technologie supérieure de Montréal, Quebec) contributed the dataset of Arabic manuscripts. The toy example (ULE) is the MNIST handwritten digit database made available by Yann LeCun and Corinna Costes.
    Object recognition (RITA) -- Antonio Torralba, Rob Fergus, and William T. Freeman, collected and made available publicly the 80 million tiny image dataset. Vinod Nair and Geoffrey Hinton collected and made available publicly the CIFAR datasets. See the techreport Learning Multiple Layers of Features from Tiny Images, by Alex Krizhevsky, 2009, for details.
    Human action recognition (HARRY) -- Ivan Laptev and Barbara Caputo collected and made publicly available the KTH human action recognition datasets. Marcin Marszałek, Ivan Laptev and Cordelia Schmid collected and made publicly available the Hollywood 2 dataset of human actions and scenes.
    Text processing (TERRY) -- David Lewis formatted and made publicly available the RCV1-v2 Text Categorization Test Collection.
    Ecology (SYLVESTER) -- Jock A. Blackard, Denis J. Dean, and Charles W. Anderson of the US Forest Service, USA, collected and made available the (Forest cover type) dataset.
  • Web platform:
    Olivier Guyon (MisterP.net, France), implemented the Causality Workbench, inspired by an earlier design of challenge platform of Steve Gunn (University of Southampton, UK).
    We are thankful to Prof. Joachim Buhmann (ETH Zurich, Switzerland) for making available the servers on which the Causality Workbench is deployed and to Thomas Fuchs (ETH Zurich, Switzerland) for administering the computer resources.
  • Project team:
    David W. Aha (Naval Research Laboratory, USA)
    Graham Taylor (New York University)
    Gideon Dror (Academic College of Tel-Aviv-Yaffo, Israel)
    Vincent Lemaire (Orange, France)
  • Advisors and beta-testers:
    We would like to thank the DARPA Deep Learning program evaluation and development teams for their input on the challenge design and the choice of the datasets. We are particularly grateful to the following researchers who reviewed the website and/or tested it.
    Gavin Cawley (University of East Anglia, UK)
    Olivier Chapelle (Yahhoo!, California)
    Ulrike von Luxburg (MPI, Germany)
    Gerard Rinkus (Brandeis University, USA)
    Quoc V. Le (Stanford University, USA)
    Danny Silver (Acadiau University, Canada)
    Grégoire Montavon (TU Berlin, Germany)

  • Coordinator:
    Isabelle Guyon
    Clopinet Enterprises
    955, Creston Road,
    Berkeley, CA 94708, U.S.A.
    Tel/Fax: (510) 524 6211
    ul @ clopinet . com