Active Learning

Simon Tong, Daphne Koller


Active Image Retrieval Demo



The standard framework in Machine Learning presents the learner with a randomly sampled data set. There has been growing interest in the area of Active Learning. Here, one allows the learner the flexibility to choose the data points that it feels are most relevant for learning a particular task. One analogy is that a standard passive learner is a student that sits and listens to a teacher while an active learner is a student that asks the teacher questions, listens to the answers and asks further questions based upon the teacher's repsonse. We are currently investigating techniques for performing active learning in three widely applicable situations: classification, density estimation and discovering causal structure. Our current results show that active learners using these techniques can outperfom regular passive learners substantially.


publications

Active Learning: Theory and Applications. (Thesis) Simon Tong. Stanford University 2001 (gzip PS)
Support Vector Machine Active Learning with Applications to Text Classification. Simon Tong, Daphne Koller. To appear Machine Learning Journal 2001. (gzip PS) (PS)
Support Vector Machine Active Learning for Image Retrieval. Simon Tong, Edward Chang. To appear in ACM Multimedia 2001. (gzip PS) (Demo)
Active Learning for Structure in Bayesian Networks. Simon Tong, Daphne Koller. International Joint Conference on Artificial Intelligence 2001. (PS)
Active Learning for Parameter Estimation in Bayesian Networks. Simon Tong, Daphne Koller. Neural Information Processing Systems 2001. (PS)
Support Vector Machine Active Learning with Applications to Text Classification. Simon Tong, Daphne Koller. Proceedings of the Seventeenth International Conference on Machine Learning. 2000. (gzip PS) (PS)



presentations

Active Learning for Structure in Bayesian Networks. Simon Tong, Daphne Koller. International Joint Conference on Artificial Intelligence 2001. (Html)
Active Learning for Parameter Estimation in Bayesian Networks. Simon Tong, Daphne Koller. Neural Information Processing Systems. 2000-2001. (Html)
Support Vector Machine Active Learning with Applications to Text Classification. Simon Tong, Daphne Koller. Proceedings of the Seventeenth International Conference on Machine Learning. 2000. (Html)