addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1light-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo

Session #9: Just do it

Active Learning

In our pervious session we touched the interesting selective learning model, wherein the classifier gained a power to  abstain (classifications). The selective learning model connects nicely with the active learning model in which the learner  may influence on the choice of its learning-examples. (Here is a theoretical result which connects the two models: pdf-link.)

Let's pick introductory level topic from the survey below. It would also be interesting to compare active learning in humans to AL in machines (the second link below).

We need session lead volunteers. Please contact me if you wanna help.

• A nice survey on active learning --

• Comparing human active learning with statistical learning predictions:

Join or login to comment.

  • Mich

    here are burr's slides i presented -
    thanks all for your ACTIVE participance.

    May 22, 2014

  • Doron

    A challanging one.

    May 21, 2014

  • Ron B.

    We had a real nice session. Thanks to Eyal G. and Eyal K. from RSA for generously hosting us. Omer gave a very interesting explanation of the magic behind word2vec method. You can read the full details here ( Send Omer a message if you wanna get the presentation slides. Yoav pointed that they also published a tech-report explaining the derivation of the word2vec model in quite detail:

    Michael lead an interesting tutorial session on the active learning model (for supervised learning). We covered a few theoretical aspects (e.g., version space, and disagreement region) and a couple of algorithmic approaches (e.g., max uncertainty sampling). Thanks!

    1 · May 21, 2014

  • Ron B.

    Hi all, I've edited the session's description.

    May 9, 2014

    • Hershel S.

      Hi Ron,

      May 11, 2014

    • Hershel S.

      Hi Ron, I'd be happy to summarize and lead a discussion of the Castro paper. Hershel

      2 · May 11, 2014

28 went

People in this
Meetup are also in:

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy