AI Paper Atlas

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

Predictive Coding beyond Gaussian Distributions

Luca Pinchetti, Tommaso Salvatori, Yordan Yordanov, Beren Millidge, Yuhang Song, Thomas Lukasiewicz

NeurIPS 2022 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Hierarchical VAEs and posterior collapse · 1 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 KL-divergence energy instead of standard Gaussian PC energyPredictive coding on M2 classification networks · MNIST and CIFAR10, M2 (softmax output)accuracy over epochsImproved
Use KL-divergence energy instead of standard Gaussian PC energyPredictive coding on M3 classification networks · MNIST and CIFAR10, M3 (softmax hidden layer)accuracy over epochsImproved
Use KL-divergence energy instead of standard Gaussian PC energyPredictive coding on M1 classification networks · MNIST and CIFAR10, M1 (tanh)accuracy over epochsNo clear effect
Use KL-divergence energy instead of standard Gaussian PC energyPredictive coding · MNISTVAE test lossNo clear effect
Use KL-divergence energy instead of standard Gaussian PC energyPredictive coding · 1B Word Benchmarktest perplexityImproved
Use KL-divergence energy instead of standard Gaussian PC energyPredictive coding · 1B Word Benchmarktest perplexityNo clear effect
Use multiple weight updates per training iterationPC_FKL transformer · 1B Word Benchmarktraining stabilityImproved
Use multiple weight updates per training iterationPC_F transformer · 1B Word Benchmarktraining stabilityHurt
Replace tanh with ReLU (ablation)Predictive coding classification networks · Classification networksaccuracyHurt
Use softmax in a hidden layer (ablation)Classification networks · M3 classification networksaccuracyMixed

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