16h - 16h40: Introduction, Isabelle Guyon, ChaLearn [slides pptx][slides pdf]
Challenges have recently proved a great stimulus for research in machine learning, pattern recognition, and robotics. Attracting hundreds of participants and the attention of a broad audience of specialists as well as sometimes the general public, these events have been important in several respects: (1) pushing the state-of-the art, (2) identifying techniques which really work, (3) attracting new researchers, (4) raising the standards of research, (5) giving the opportunity to non-established researchers to make themselves rapidly known.
Since 2003, we have been organizing challenges in machine learning and data science and have recently incorporated a non-profit organization (ChaLearn) making available resources to future challenge organizers. Challenges are a great means of carrying out scientific research by focusing the mental energy of brilliant people around the world willing to donate their free time to tackle interesting problems. But, what makes a good challenge that provides conclusive results having an important impact? This introduction will survey the landscape of challenges, present some highlights of recent challenges, and point to new directions.
16h40 - 17h30: Panel discussion
Do you want to organize your own challenge? We invited challenge organizers to answer your questions. See our list of panelists below.
Panelists:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Florence d'Alché-Buc, Professor at Télécom ParisTech
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Florence d'Alché-Buc is a full professor at Télécom ParisTech. She has initiated and chaired the Challenge program of the european network of excellence PASCAL, which gave rise to many recurrent challenges (Visual Object Classes, Text entailment, ...). She participated in several bioinformatics challenges including DREAM challenges, which she uses as par of her classes on biological network inference.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Balázs Kégl, CNRS and Paris-Saclay Center for Data Science
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Balázs Kégl is a CR1 researcher at LAL (Linear Accelerator Laboratory), working on establishing a multi-disciplinary research team that links state-of-the-art statistical, machine learning, and signal processing methodologies to applications in experimental physics. He is leading the Paris-Saclay Center for Data Science federating research in data science between a dozen of universities and engineering schools. He is a co-organizer of the Higgs Boson challenge, a highly successful challenge ran in 2014, which attracted 1785 teams [see slides].
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Michèle Sebag, LRI, France : Pascal challenges
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Michele Sebag is senior scientist at the CNRS and co-head of project TAO, INRIA Saclay, France. She has been heading the challenge program of the Pascal European network of excellence, which has launched many high impact challenges, including challenges in text mining, brain computer interfaces, and computer vision (Pascal VOC challenges). She will share her vision on where challenges in machine learning should head.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Sergio Escalera, University of Barcelona, Spain
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Dr. Sergio Escalera leads the Human Pose Recovery and Behavior Analysis Group at UB, CVC, and the Barcelona Graduate School of Mathematics. He is an associate professor at the Department of Applied Mathematics and Analysis, Universitat de Barcelona and a part time professor at Universitat Oberta de Catalunya. After winning a prize in the 2012 ChaLearn gesture recognition challenge sponsored in part by Microsoft Kinect, he has been leading the organizing team of the part two ChaLearn Looking at people challenges in 2013 and 2014.
Coordinatrice:
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Isabelle Guyon, ChaLearn, Berkeley, California
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Isabelle Guyon is an independent consultant, specialized in statistical data analysis, pattern recognition and machine learning. Prior to starting her consulting practice in 1996, Isabelle Guyon was a researcher at AT&T Bell Laboratories, where she pioneered applications of neural networks to pen computer interfaces and co-invented Support Vector Machines (SVM). She is also the primary inventor of SVM-RFE, a variable selection technique based on SVM. She organized many challenges in Machine Learning over the past few years supported by the EU network Pascal2, NSF, and DARPA, with prizes sponsored by Microsoft, Google, and Texas Instrument. She is president of Chalearn, a non-profit dedicated to organizing challenges, vice-president of the Unipen foundation, adjunct professor at New-York University, action editor of the Journal of Machine Learning Research, and editor of the Challenges in Machine Learning book series of Microtome.