Laboratory for Intelligent Probabilistic Systems

One approach to AI research is to work directly on applications that matter — say, trying to improve production systems for speech recognition or medical imaging. But most research, even in applied fields like computer vision, is done on highly simplified proxies for the real world. Progress on object recognition benchmarks — from toy-ish ones like MNIST, NORB, and Caltech101, to complex and challenging ones like ImageNet and Pascal VOC — isn’t valuable in its own right, but only insofar as it yields insights that help us design better systems for real applications. So it’s natural to ask: which research results will generalize to new situations?

When we talk about priors and regularization, we often motivate them in terms of “incorporating knowledge” or “preventing overfitting.” In a sense, the two are equivalent: any prior or regularizer must favor certain explanations relative to others, so favoring one explanation is equivalent to punishing others. But I’ll argue that these are two very different phenomena, and it’s useful to know which one is going on.

In this post, I’ll summarize one of my favorite papers from ICML 2013: Fast Dropout Training, by Sida Wang and Christopher Manning. This paper derives an analytic approximation to dropout, a randomized regularization method recently proposed for training deep nets that has allowed big improvements in predictive accuracy.   Their approximation gives a roughly 10-times speedup under certain conditions.  Much more interestingly, the authors also show strong connections to existing regularization methods, shedding light on why dropout works so well.