Travelled to:1 × Finland
1 × United Kingdom
2 × Canada
3 × USA
Collaborated with:O.Shamir E.Hazan A.Tewari R.Arora D.Vainsencher S.Mannor J.Keshet Y.Singer J.Ding T.Koren Y.Peres R.Gilad-Bachrach L.Xiao
Talks about:regret (3) onlin (3) adversari (2) bandit (2) learn (2) determinist (1) distribut (1) hierarch (1) function (1) classifi (1)
Person: Ofer Dekel
DBLP: Dekel:Ofer
Contributed to:
2014
201320122011200920082004Wrote 8 papers:
- STOC-2014-DekelDKP
- Bandits with switching costs: T2/3 regret (OD, JD, TK, YP), pp. 459–467.
- ICML-c3-2013-DekelH
- Better Rates for Any Adversarial Deterministic MDP (OD, EH), pp. 675–683.
- ICML-2012-DekelTA #adaptation #learning #online #policy
- Online Bandit Learning against an Adaptive Adversary: from Regret to Policy Regret (OD, AT, RA), p. 227.
- ICML-2011-DekelGSX #distributed #online #predict
- Optimal Distributed Online Prediction (OD, RGB, OS, LX), pp. 713–720.
- ICML-2011-VainsencherDM #estimation #online
- Bundle Selling by Online Estimation of Valuation Functions (DV, OD, SM), pp. 1137–1144.
- ICML-2009-DekelS #education
- Good learners for evil teachers (OD, OS), pp. 233–240.
- ICML-2008-DekelS #learning
- Learning to classify with missing and corrupted features (OD, OS), pp. 216–223.
- ICML-2004-DekelKS #classification #scalability
- Large margin hierarchical classification (OD, JK, YS).