AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

Discretely Relaxing Continuous Variables for tractable Variational Inference

Trefor Evans, Prasanth Nair

NeurIPS 2018 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Variational inference theory and gradients · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use exact DIRECT ELBO gradients instead of REINFORCE gradient estimatesdiscrete-prior GLM regression · synthetic random Fourier feature regressionoptimization iterations and wall-clock timeImproved
Discretely relax a Gaussian prior and train with DIRECT mean-field inference instead of a continuous Gaussian prior with reparameterizationGLM regression · UCI regression datasetstest RMSEMixed
Discretely relax a Gaussian prior and train with DIRECT mean-field inference instead of a continuous Gaussian prior with reparameterizationGLM regression · UCI regression datasetstraining timeImproved
Use a 5-component mixture variational distribution instead of mean-field (ablation)DIRECT GLM regression · UCI regression datasetstest RMSEImproved
Train the mixture variational distribution using an ELBO lower bound instead of unbiased SGD (ablation)DIRECT mixture GLM regression · UCI regression datasetstest RMSEHurt
Train the mixture variational distribution using an ELBO lower bound instead of unbiased SGD (ablation)DIRECT mixture GLM regression · UCI regression datasetstraining timeImproved
Use a mixture variational distribution instead of mean-field (ablation)DIRECT GLM regression · UCI regression datasetstraining timeHurt
Represent posterior samples using discrete low-precision valuesDIRECT GLM regression · Gasposterior sample sparsity: over 84%Improved
Represent posterior samples using discrete low-precision valuesDIRECT GLM regression · Electricposterior sample sparsity: over 99.6%Improved

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