Travelled to:1 × China
1 × France
3 × USA
Collaborated with:S.Agarwal P.Kar P.Jain S.Satheesh Pramala Senthil H.G.Ramaswamy A.Saha A.Agarwal S.Kalyanakrishnan A.K.Menon S.Chawla
Talks about:class (4) optim (3) base (3) algorithm (2) perform (2) partial (2) consist (2) measur (2) binari (2) auc (2)
Person: Harikrishna Narasimhan
DBLP: Narasimhan:Harikrishna
Contributed to:
2015201420132013
2013
2009Wrote 8 papers:
- ICML-2015-KarN0 #precise
- Surrogate Functions for Maximizing Precision at the Top (PK, HN, PJ), pp. 189–198.
- ICML-2015-NarasimhanK0 #metric #optimisation #performance
- Optimizing Non-decomposable Performance Measures: A Tale of Two Classes (HN, PK, PJ), pp. 199–208.
- ICML-2015-NarasimhanRS0 #algorithm #consistency #metric #multi #performance
- Consistent Multiclass Algorithms for Complex Performance Measures (HN, HGR, AS, SA), pp. 2398–2407.
- ICML-c2-2014-0001NKA #estimation #probability
- GEV-Canonical Regression for Accurate Binary Class Probability Estimation when One Class is Rare (AA, HN, SK, SA), pp. 1989–1997.
- ICML-c1-2013-NarasimhanA #approach #optimisation
- A Structural SVM Based Approach for Optimizing Partial AUC (HN, SA), pp. 516–524.
- ICML-c3-2013-MenonNAC #algorithm #classification #consistency #on the #statistics
- On the Statistical Consistency of Algorithms for Binary Classification under Class Imbalance (AKM, HN, SA, SC), pp. 603–611.
- KDD-2013-NarasimhanA #bound #named #optimisation
- SVMpAUCtight: a new support vector method for optimizing partial AUC based on a tight convex upper bound (HN, SA), pp. 167–175.
- FDG-2009-SatheeshNS #evolution #game studies
- Evolving player-specific content for level based arcade games (SS, HN, PS), pp. 329–330.