Research map / Probabilistic and scientific modeling / Dynamical systems
Neural Networks for Efficient Bayesian Decoding of Natural Images from Retinal Neurons
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| convolutional autoencoder enhancement of linear decoded image (L-CAE) | linear decoder · ImageNet test set | MSE: 34.50% MSE reduction | Helped |
| replace CAE with a 6th degree polynomial static nonlinearity (ablation) | linear decoder · ImageNet test set | MSE: 3.25% MSE reduction (much less than L-CAE) | Helped |
| training linear decoder and L-CAE on CelebA faces instead of ImageNet (ablation) | linear decoder · CelebA test set | MSE: 17% lower average test MSE | Helped |
| training L-CAE on CelebA faces instead of ImageNet (ablation) | L-CAE · CelebA test set | MSE: 5% lower MSE; smaller gap than linear decoder | Helped |
| training on phase-scrambled natural images instead of natural images (ablation) | L-CAE · ImageNet test set | MSE | Hurt |
| training on phase-scrambled natural images instead of natural images (ablation) | linear decoder · ImageNet test set | MSE | No clear effect |
| perceptual VGG-based loss instead of MSE (ablation) | L-CAE · ImageNet | perceptual quality | No clear effect |
| increasing CAE training data from 20k to around 500k examples (ablation) | L-CAE · ImageNet | MSE | Helped |
Closest papers by meaning
- Maximum a posteriori natural scene reconstruction from retinal ganglion cells with deep denoiser priors (NeurIPS 2022 · similarity 0.91)
- A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks (NeurIPS 2025 · similarity 0.87)
- Continuous Spatiotemporal Events Decoupling through Spike-based Bayesian Computation (NeurIPS 2024 · similarity 0.87)
- MonkeySee: Space-time-resolved reconstructions of natural images from macaque multi-unit activity (NeurIPS 2024 · similarity 0.86)
- Finding Visual Saliency in Continuous Spike Stream (AAAI 2024 · similarity 0.85)
- Decomposing stimulus-specific sensory neural information via diffusion models (NeurIPS 2025 · similarity 0.85)
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.