Research map / Architectures and efficiency / CNN design
Deep Learning Models of the Retinal Response to Natural Scenes
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 |
|---|---|---|---|
| replace LN/GLM encoding models with 3-layer CNN | retinal encoding model · salamander RGC natural scenes and white noise | correlation coefficient | Helped |
| increase CNN depth beyond three layers (ablation) | CNN retina model · held-out retinal responses | prediction accuracy | No clear effect |
| use large convolutional filters (>15x15) instead of small (ablation) | CNN retina model · held-out retinal responses | prediction accuracy | Helped |
| add recurrent layer (LSTM/GRU/MUT) after CNN (ablation) | CNN retina model · natural scenes | prediction accuracy | No clear effect |
| inject noise during training (ablation) | CNN retina model · retinal response variability | sub-Poisson noise scaling | Helped |
Closest papers by meaning
- From deep learning to mechanistic understanding in neuroscience: the structure of retinal prediction (NeurIPS 2019 · similarity 0.93)
- Natural image synthesis for the retina with variational information bottleneck representation (NeurIPS 2022 · similarity 0.92)
- A Unified Theory of Early Visual Representations from Retina to Cortex through Anatomically Constrained Deep CNNs (ICLR 2019 · similarity 0.92)
- Maximum a posteriori natural scene reconstruction from retinal ganglion cells with deep denoiser priors (NeurIPS 2022 · similarity 0.91)
- Finding Visual Saliency in Continuous Spike Stream (AAAI 2024 · similarity 0.88)
- Recurrent neural network dynamical systems for biological vision (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.