Skip to yearly menu bar Skip to main content


Dropout as a Structured Shrinkage Prior

Eric Nalisnick · Jose Miguel Hernandez-Lobato · Padhraic Smyth

Pacific Ballroom #84

Keywords: [ Monte Carlo Methods ] [ Bayesian Methods ] [ Bayesian Deep Learning ] [ Approximate Inference ]


Dropout regularization of deep neural networks has been a mysterious yet effective tool to prevent overfitting. Explanations for its success range from the prevention of "co-adapted" weights to it being a form of cheap Bayesian inference. We propose a novel framework for understanding multiplicative noise in neural networks, considering continuous distributions as well as Bernoulli noise (i.e. dropout). We show that multiplicative noise induces structured shrinkage priors on a network's weights. We derive the equivalence through reparametrization properties of scale mixtures and without invoking any approximations. Given the equivalence, we then show that dropout's Monte Carlo training objective approximates marginal MAP estimation. We leverage these insights to propose a novel shrinkage framework for resnets, terming the prior 'automatic depth determination' as it is the natural analog of automatic relevance determination for network depth. Lastly, we investigate two inference strategies that improve upon the aforementioned MAP approximation in regression benchmarks.

Live content is unavailable. Log in and register to view live content