12th Annual Conference of the International Speech Communication Association

Florence, Italy
August 27-31. 2011

Large Margin - Minimum Classification Error Using Sum of Shifted Sigmoids as the Loss Function

Madhavi V. Ratnagiri (1), Biing-Hwang Juang (2), Lawrence Rabiner (1)

(1) Rutgers University, USA
(2) Georgia Institute of Technology, USA

Mon-Ses1-P4.13, #13 We have developed a novel loss function that embeds large-margin classification into Minimum Classification Error (MCE) training. Unlike previous efforts this approach employs a loss function that is bounded, does not require incremental adjustment of the margin or prior MCE training. It extends the Bayes risk formulation of MCE using Parzen Window estimation to incorporate large-margin classification and develops a loss function that is a sum of shifted sigmoids. Experimental results show improvement in recognition performance when evaluated on the TIDigits database.

Full Paper

Bibliographic reference.  Ratnagiri, Madhavi V. / Juang, Biing-Hwang / Rabiner, Lawrence (2011): "Large margin - minimum classification error using sum of shifted sigmoids as the loss function", In INTERSPEECH-2011, 1729-1732.