[an error occurred while processing this directive] [an error occurred while processing this directive]
Po-Sen Huang

Shallow and deep learning for audio and natural language processing

Po-Sen Huang, 9/23/2014, 4:30-5:30pm, BI 2369

Machine learning can be viewed as the problem of finding the optimum hypothesis in a pre-determined hypothesis space given the input data and targets. Ideally, the optimum hypothesis can be approached by models with high expressive power given enough data. In this talk, I explore the problem of finding an optimum hypothesis for audio and natural language processing from three different aspects: deep learning, joint deep learning with kernel machines, and shallow kernel machines.

[an error occurred while processing this directive]