My depth oral is past (and passed), but with reservation. My committee expressed disappointment with what I presented to them; they felt that I did not show either a sufficient mastery of the material upon which I planned to do original research (develop new methods rather than compare and consume or reuse old methods), nor did I present anything exciting for new research directions. This means I’ve got to focus on delivering a really promising progress report in six months time.
What’s going to make up that report? A review of how distance metric learning is performed (Learning Bregman Distance Functions by Lei Wu et al., Distance Metric Learning, with Application to Clustering with Side-Information by Eric Xing for starters), along with the results of applying those methods to different formulations of semi-supervised learning.
A busy six months ahead 🙂