Sampling-free variational inference of bayesian neural networks by variance backpropagation
dc.contributor.author | Haußmann, M. | |
dc.contributor.author | Hamprecht, F. A. | |
dc.contributor.author | Kandemir, Melih | |
dc.date.accessioned | 2024-03-08T08:41:16Z | |
dc.date.available | 2024-03-08T08:41:16Z | |
dc.date.issued | 2020 | |
dc.identifier.issn | 2640-3498 | en_US |
dc.identifier.uri | http://hdl.handle.net/10679/9280 | |
dc.identifier.uri | https://proceedings.mlr.press/v115/haussmann20a.html | |
dc.description.abstract | We propose a new Bayesian Neural Net formulation that affords variational inference for which the evidence lower bound is analytically tractable subject to a tight approximation. We achieve this tractability by (i) decomposing ReLU nonlinearities into the product of an identity and a Heaviside step function, (ii) introducing a separate path that decomposes the neural net expectation from its variance. We demonstrate formally that introducing separate latent binary variables to the activations allows representing the neural network likelihood as a chain of linear operations. Performing variational inference on this construction enables a sampling-free computation of the evidence lower bound which is a more effective approximation than the widely applied Monte Carlo sampling and CLT related techniques. We evaluate the model on a range of regression and classification tasks against BNN inference alternatives, showing competitive or improved performance over the current state-of-the-art. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | ML Research Press | en_US |
dc.relation.ispartof | Proceedings of Machine Learning Research | |
dc.rights | restrictedAccess | |
dc.title | Sampling-free variational inference of bayesian neural networks by variance backpropagation | en_US |
dc.type | Conference paper | en_US |
dc.publicationstatus | Published | en_US |
dc.contributor.department | Özyeğin University | |
dc.contributor.authorID | (ORCID 0000-0001-6293-3656 & YÖK ID 258737) Kandemir, Melih | |
dc.contributor.ozuauthor | Kandemir, Melih | |
dc.identifier.volume | 115 | en_US |
dc.identifier.startpage | 563 | en_US |
dc.identifier.endpage | 573 | en_US |
dc.identifier.wos | WOS:000722423500051 | |
dc.identifier.scopus | SCOPUS:2-s2.0-85162227302 | |
dc.relation.publicationcategory | Conference Paper - International - Institutional Academic Staff |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
Share this page