Research map / Architectures and efficiency / CNN design
Biologically inspired CNN explanation
Explains CNN behavior and links networks to human visual cortex and perceptual learning.
46 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in CNN design, keyword cluster “cnn · filter · trained”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
5 papers in 2023–24 and 11 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 3.6% to 10.8%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Eye Tracking for Everyone (CVPR 2016 · 934 citations)
- 2017 · Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer (ICLR 2017 · 1,535 citations)
- 2018 · Matrix capsules with EM routing (ICLR 2018 · 726 citations)
- 2019 · Mixed Effects Neural Networks (MeNets) With Applications to Gaze Estimation (CVPR 2019 · 102 citations)
- 2020 · On the Relationship between Self-Attention and Convolutional Layers (ICLR 2020 · 89 citations)
- 2021 · Relevance-CAM: Your Model Already Knows Where To Look (CVPR 2021 · 78 citations)
- 2022 · AutoMTL: A Programming Framework for Automating Efficient Multi-Task Learning (NeurIPS 2022 · 5 citations)
- 2023 · BioNet: A Biologically-Inspired Network for Face Recognition (CVPR 2023 · 7 citations)
- 2024 · MonkeySee: Space-time-resolved reconstructions of natural images from macaque multi-unit activity (NeurIPS 2024 · 2 citations)
- 2025 · ParseCaps: An Interpretable Parsing Capsule Network for Medical Image Diagnosis (AAAI 2025 · 2 citations)
- 2026 · Connecting Independently Trained Modes via Layer-Wise Connectivity (ICML 2026 · 0 citations)
Earliest papers here
- Eye Tracking for Everyone (CVPR 2016 · 934 citations)
- Deep Learning Models of the Retinal Response to Natural Scenes (NeurIPS 2016 · 186 citations)
- Recurrently Target-Attending Tracking (CVPR 2016 · 148 citations)
What has been tried
37 papers here report 371 measured changes (a component added, removed or swapped, compared with a baseline): helped 68%, no clear effect 11%, hurt 18%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add triplet loss L_t | PH² · ACC | 95.01 -> 97.53 | Helped | ParseCaps: An Interpretable Parsing Capsule Network for Medical Image Diagnosis (AAAI 2025 · 2 citations) |
| Add P-cell pathway | Robustness benchmark (reduced attack set) · Robustness Score | 38.4 -> 44.6 | Helped | Explicitly Modeling Subcortical Vision with a Neuro-Inspired Front-End Improves CNN Robustness (NeurIPS 2025 · 1 citation) |
| Omit covariance term tr(AΣ) | CIFAR-10 · accuracy | 93.17% -> 37.78% | Hurt | Dendritic Integration Inspired Artificial Neural Networks Capture Data Correlation (NeurIPS 2024 · 1 citation) |
| Use later rather than earlier ResNet-18 layers | Odd-one-out task, most GT classes · average rank | Helped | Alignment of CNN and Human Judgments of Geometric and Topological Concepts (AAAI 2025 · 0 citations) | |
| initialize the data-driven model with pretrained parameters rather than random parameters | NSD · predictive accuracy | Helped | A Convolutional Neural Network Interpretable Framework for Human Ventral Visual Pathway Representation (AAAI 2024) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 16 papers; from OpenAlex and the papers' affiliations).
- Shanghai Jiao Tong University · 2 papers
Use it
Download BibTeX of 34 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Filter design and trained-filter analysis (128 papers)
- Compact CNN architectures (114 papers)
- Theory of convolutional inductive bias (64 papers)
- CNN filter pruning (43 papers)
Elsewhere in the topic
- Efficient CNN scaling and dynamic inference (104 papers)
- Dense matching and estimation CNNs (93 papers)
- Invertible and wide-network analysis (65 papers)
- Large and dynamic kernel design (62 papers)
- Group symmetry equivariant networks (55 papers)
- Steerable rotation-equivariant CNNs (40 papers)
- Higher-order pooling (20 papers)
- Group convolution theory (19 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).