11 Aug 2014

Lab meeting on Wednesday, 13th of Aug, will be divided into two parts. First, Anoop will give a short talk on this paper:

Kneser-Ney Smoothing on Expected Counts. by Hui Zhang and David Chiang. ACL 2014

Widely used in speech and language processing, Kneser-Ney (KN) smoothing has consistently been shown to be one of the best-performing smoothing methods. However, KN smoothing assumes integer counts. In this paper, we propose a generalization of KN smoothing that operates on fractional counts, or, more precisely, on distributions over counts. We rederive all the steps of KN smoothing to operate on count distributions instead of integral counts.

And then Milan will give a quick overview of discourse and continue his talk from last week.The meeting will be at TASC1 9408 from 1100 hours.