Active Learning Challenge Isabelle Guyon (Clopinet, California) Gavin Cawley (University of East Anglia, UK) Olivier Chapelle (Yahhoo!, California) Gideon Dror (Academic College of Tel-Aviv-Yaffo, Israel) Vincent Lemaire (Orange, France) Amir Reza Saffari Azar (Graz University of Technology) Alexander Statnikov (New York University, USA) Active Learning Challenge http: //clopinet. com/al
What is the problem? Active Learning Challenge http: //clopinet. com/al
Labeling data is expensive $$ Active Learning Challenge $$$$$ http: //clopinet. com/al
Examples of domains • • Chemo-informatics Handwriting and speech recognition Image processing Text processing Marketing Ecology Embryology Active Learning Challenge http: //clopinet. com/al
What is active learning? Active Learning Challenge http: //clopinet. com/al
What is out there? Active Learning Challenge http: //clopinet. com/al
Scenarios Burr Settles. Active Learning Literature Survey. CDTR 1648, Univ. Wisconsin–Madison. http: //clopinet. com/al Active Learning Challenge
“De novo” queries De novo queries implicitly assume interventions on the system under study: not for this challenge Active Learning Challenge http: //clopinet. com/al
Focus on “pool-based” AL • Simplest scenario for a challenge. Training data: labels can be queried Test data: unknown labels • Methods developed for pool-based AL should also be useful for stream-based AL. Active Learning Challenge http: //clopinet. com/al
Example Accuracy=0. 7 Accuracy=0. 9 (a) Toy 2 -class problem, 400 instances Gaussian distributed. (b) Linear logistic regression model trained w. 30 random instance (c) Linear logistic regression model trained w. 30 actively queried instances using “uncertainty sampling”. Active Learning Challenge Burr Settles, 2009 http: //clopinet. com/al
Learning curve Active Learning Challenge Burr Settles, 2009 http: //clopinet. com/al
Other methods • Expected model change (greatest gradient if sample were used for training) • Query by committee (query the sample subject to largest disagreement) • Bayesian active learning (maximize change in revised posterior distribution) • Expected error reduction (maximize generalization performance improvement) • Information density (ask for examples both informative and representative) Active Learning Challenge Burr Settles, 2009 http: //clopinet. com/al
Datasets Active Learning Challenge http: //clopinet. com/al
Data donors This project would not have been possible without generous donations of data: • Chemoinformatics -- Charles Bergeron, Kristin Bennett and Curt Breneman (Rensselaer Polytechnic Institute, New York) contributed a dataset, which will be used for final testing. • Embryology -- Emmanuel Faure, Thierry Savy, Louise Duloquin, Miguel Luengo Oroz, Benoit Lombardot, Camilo Melani, Paul Bourgine, and Nadine Peyriéras (Institut des systèmes complexes, France) contributed the ZEBRA dataset. • Handwriting recognition -- Reza Farrahi Moghaddam, Mathias Adankon, Kostyantyn Filonenko, Robert Wisnovsky, and Mohamed Chériet (Ecole de technologie supérieure de Montréal, Quebec) contributed the IBN_SINA dataset. • Marketing -- Vincent Lemaire, Marc Boullé, Fabrice Clérot, Raphael Féraud, Aurélie Le Cam, and Pascal Gouzien (Orange, France) contributed the ORANGE dataset, previously used in the KDD cup 2009. We also reused data made publicly available on the Internet: • Chemoinformatics -- The National Cancer Institute (USA) for the HIVA dataset. • Ecology -- Jock A. Blackard, Denis J. Dean, and Charles W. Anderson (US Forest Service, USA) for the SYLVA dataset (Forest cover type). • Text processing -- Tom Mitchell (USA) and Ron Bekkerman (Israel) for the NOVA datset (derived from the Twenty Newsgroups). Active Learning Challenge http: //clopinet. com/al
Development datasets Active Learning Challenge http: //clopinet. com/al
Difficulties • • • Spase data Missing values Unbalanced classes Categorical variables Noisy data Large datasets Active Learning Challenge http: //clopinet. com/al
Final test datasets • Will serve to do the final ranking • Will be from the same domains • May have different data representations and distributions • No feed-back: the results will not be revealed until the end of the challenge Active Learning Challenge http: //clopinet. com/al
Protocol Active Learning Challenge http: //clopinet. com/al
Virtual Lab Virtual cash Active Learning Challenge Joint work with: • Constantin Aliferis, New York University • Gregory F. Cooper, Pittsburg University • André Elisseeff, Nhumi, Zürich • Jean-Philippe Pellet, IBM Zürich • Alexander Statnikov, New York University • Peter Spirtes, Carnegie Mellon http: //clopinet. com/al
Step by step instructions Download the data. You get 1 labeled example. 1. Predict 2. Sample 3. Submit a query 4. Retrieve the labels Active Learning Challenge http: //clopinet. com/al
Two phases • Development phase: – 6 datasets available – Can try as many times as you want – Matlab users can run queries on their computers – Others can use the labels (provided) • Final test phase: – 6 new datasets available – A single try – No feed-back Active Learning Challenge http: //clopinet. com/al
Evaluation Active Learning Challenge http: //clopinet. com/al
AUC score For each set of samples queried, we assess the predictions of the learning machine with the Area under the ROC curve. Active Learning Challenge http: //clopinet. com/al
Area under the Learning Curve (ALC) Linear interpolation. Horizontal extrapolation. One query Five queries Thirteen queries Lazy: ask for all labels at once Active Learning Challenge http: //clopinet. com/al
Prizes If you win on… • • 1 dataset: $100 2 datasets: $200 3 datasets: $400 4 datasets: $800 5 datasets: $1600 6 datasets: $3200! Plus travel awards for top ranking students. Active Learning Challenge http: //clopinet. com/al
Schedule Active Learning Challenge http: //clopinet. com/al
Conclusion Try our new challenge, learn, and win!!!! – Workshops: • AISTATS 2010, Sardinia, May, 2010 • WCCI 2010 Workshop, Barcelona, July, 2010 • Travel awards for top ranking students. – Proceedings published by JMLR & IEEE. – Prizes: P(i)=$100 * 2(n-1) – Your problem solved by dozens of research groups: • Help us organize the next challenge! Active Learning Challenge http: //clopinet. com/al
Скачать презентацию Active Learning Challenge Isabelle Guyon Clopinet California Gavin
7a3dadbcd97a34354154f1a5804e0082.ppt