1 / 22

Active Learning

Active Learning. Meeting 5 — October 21, 2014 CSCE 6933 Rodney Nielsen. Active Learning. Usually an abundance of unlabeled data How much should you label? Which instances should you label? Does it matter? Can the learner benefit from selective labeling?

hisoki
Download Presentation

Active Learning

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Active Learning Meeting 5 — October 21, 2014 CSCE 6933 Rodney Nielsen

  2. Active Learning • Usually an abundance of unlabeled data • How much should you label? • Which instances should you label? • Does it matter? • Can the learner benefit from selective labeling? • Active Learning: incrementally request labels for instances believed to be informative

  3. Learning Paradigms random ? query ? ? ? ? random Supervised Learning Unsupervised Learning Active Learning

  4. Active Learning Applications • Speech Recognition • 10 mins to annotate words in 1 min of speech • 7 hrs to annotate phonemes of 1 minute speech • Named Entity Recognition • Half an hour for a simple newswire article • PhD for a bioinformatics article • Image annotation

  5. Heuristic Active Learning Algorithm

  6. Heuristic Active Learning Algorithm • Start with unlabeled data • Randomly pick small num exs to have labeled • Repeat • Train classifier on labeled data • Query the unlabeled ex that: • Is closest to the boundary • Has the least certainty • Minimizes overall uncertainty random ? query random ? ? ? ?

  7. Active Learning Performance Ex. • Document classification: baseball vs. hockey

  8. Space of Active Learning

  9. Space of Active Learning

  10. Active Learning Query Types

  11. Membership Query Synthesis • Dynamically construct query instances based on expected informativeness • Applications • Character recognition. • Robot scientist: find optimal growth medium for a yeast • 3x $ decrease vs. cheapest next • 100x $ decrease vs. random selection

  12. Questions • Membership Query Synthesis

  13. Stream-based Selective Sampling • Informativeness measure • Region of uncertainty / Version space • Applications • POST • Sensor scheduling • IR ranking • WSD

  14. Pool-based Active Learning • Informativeness measure • Applications • Cancer diagnosis • Text classification • IE • Image classfctn & retrieval • Video classfctn & retrieval • Speech recognition

  15. Pool-based Active Learning Loop

  16. Space of Active Learning

  17. Questions • Questions???

  18. Uncertainty Sampling • Uncertainty sampling • Select examples based on confidence in prediction • Probabilistic models • Entropy-based models

  19. Questions • Entropy

  20. Entropy

  21. Heat Map

  22. Instance Selection Methods • Informativeness measures • Region of uncertainty • Information density

More Related