Linear Factor Models are used to recover the latent variables from observers, allowing us to discover explanatory factors, and to understand better about how machine thinks. There are several variants of Linear Factor Models, such as Probabilistic PCA, Independent Component Analysis, Slow Feature Analysis, and Sparse Coding. Chapter 15 (LFM) covers goals, construction methods, some extension and applications of these models. Speaker(s): Xinye Xu (University of Michigan)
Building: | East Hall |
---|---|
Event Type: | Workshop / Seminar |
Tags: | Mathematics |
Source: | Happening @ Michigan from Department of Mathematics |