Research map / Architectures and efficiency / RNNs and normalization
Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs
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 |
|---|---|---|---|
| Add recurrent steps to the blocks (ablation) | CORnet-S · Brain-Score and ImageNet | Brain-Score and ImageNet top-1 performance | Helped |
| Use a bottleneck with at least 4x expansion (ablation) | CORnet-S · CORnet-S circuitry analysis | Brain-Score | Helped |
| Add a skip connection (ablation) | CORnet-S · CORnet-S circuitry analysis | Brain-Score | Helped |
| Add more recurrent steps (ablation) | CORnet-S · CORnet-S circuitry analysis | Brain-Score | No clear effect |
| Add a fifth area (ablation) | CORnet-S · CORnet-S circuitry analysis | Brain-Score | No clear effect |
| Use two rather than three convolutions per block (ablation) | CORnet-S · CORnet-S circuitry analysis | ImageNet top-1 performance | Mixed |
| Increase the number of areas (ablation) | CORnet-S · CORnet-S circuitry analysis | ImageNet top-1 performance | Mixed |
| Use batch normalization per time step rather than global group normalization (ablation) | CORnet-S · CORnet-S circuitry analysis | ImageNet top-1 performance | Mixed |
| Change the type of gating (ablation) | CORnet-S · CORnet-S circuitry analysis | Brain-Score | No clear effect |
| Train for additional epochs rather than selecting the epoch with maximum Brain-Score | CORnet-S · ImageNet | top-1 accuracy: 73.1\% -> 74.4\% | Helped |
Closest papers by meaning
- Task-Driven Convolutional Recurrent Models of the Visual System (NeurIPS 2018 · similarity 0.90)
- Scaling Laws for Task-Optimized Models of the Primate Visual Ventral Stream (ICML 2025 · similarity 0.89)
- TopoNets: High performing vision and language models with brain-like topography (ICLR 2025 · similarity 0.88)
- Vision CNNs trained to estimate spatial latents learned similar ventral-stream-aligned representations (ICLR 2025 · similarity 0.87)
- STSBench: A Large-Scale Dataset for Modeling Neuronal Activity in the Dorsal Stream of Primate Visual Cortex (NeurIPS 2025 · similarity 0.87)
- Explicitly Modeling Subcortical Vision with a Neuro-Inspired Front-End Improves CNN Robustness (NeurIPS 2025 · 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.