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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding on M2 classification networks · MNIST and CIFAR10, M2 (softmax output) | accuracy over epochs | Improved |
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding on M3 classification networks · MNIST and CIFAR10, M3 (softmax hidden layer) | accuracy over epochs | Improved |
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding on M1 classification networks · MNIST and CIFAR10, M1 (tanh) | accuracy over epochs | No clear effect |
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding · MNIST | VAE test loss | No clear effect |
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding · 1B Word Benchmark | test perplexity | Improved |
| Use KL-divergence energy instead of standard Gaussian PC energy | Predictive coding · 1B Word Benchmark | test perplexity | No clear effect |
| Use multiple weight updates per training iteration | PC_FKL transformer · 1B Word Benchmark | training stability | Improved |
| Use multiple weight updates per training iteration | PC_F transformer · 1B Word Benchmark | training stability | Hurt |
| Replace tanh with ReLU (ablation) | Predictive coding classification networks · Classification networks | accuracy | Hurt |
| Use softmax in a hidden layer (ablation) | Classification networks · M3 classification networks | accuracy | Mixed |
Closest papers by meaning
- Divide-and-Conquer Predictive Coding: a structured Bayesian inference algorithm (NeurIPS 2024 · similarity 0.88)
- A Stable, Fast, and Fully Automatic Learning Algorithm for Predictive Coding Networks (ICLR 2024 · similarity 0.88)
- Sample as you Infer: Predictive Coding with Langevin Dynamics (ICML 2024 · similarity 0.88)
- $\mu$PC: Scaling Predictive Coding to 100+ Layer Networks (NeurIPS 2025 · similarity 0.87)
- Understanding and Improving Optimization in Predictive Coding Networks (AAAI 2024 · similarity 0.87)
- Only Strict Saddles in the Energy Landscape of Predictive Coding Networks? (NeurIPS 2024 · similarity 0.87)
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.
