AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Implicit Neural Representation Inference for Low-Dimensional Bayesian Deep Learning

Panagiotis Dimitrakopoulos, Giorgos Sfikas, Christophoros Nikou

ICLR 2024 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Variational Bayesian deep learning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes factoring deep network parameters into deterministic and probabilistic components, with the probabilistic part inferred in a low-dimensional latent space via an implicit neural representation, to make Bayesian deep learning tractable.

Problem

Bayesian inference for overparameterized deep networks requires severe approximations that sacrifice predictive strength, while providing well-calibrated full predictive distributions.

Approach

Factor model parameters as a function of deterministic and probabilistic components so that Bayesian inference is performed over a low-dimensional probabilistic latent, with an implicit neural representation mapping it to the full parameter set.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy