Research map / Vision-language and multimodal / Human and embodied modeling
Neuro-Inspired Visual Pattern Recognition via Biological Reservoir Computing
Luca Ciampi, Ludovico Iannello, Fabrizio Tonelli, Gabriele Lagani, Angelo Di Garbo, Federico Cremisi, Giuseppe M. Amato
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 |
|---|---|---|---|
| shorten post-stimulus readout window W to 5 ms (ablation) | BRC readout · Clock-digit-like patterns | accuracy | Improved |
| use readout window W of 10 ms instead of longer windows (ablation) | BRC readout · MNIST subset (200 images) | accuracy | Improved |
| train readout on Day 1 and test on later day (cross-day drift) (ablation) | BRC · Clock-digit-like patterns | accuracy: 46 ± 15% on Day 2 and 37 ± 19% on Day 3 vs chance 12% | Hurt |
| train readout on Day 1 and test on later day (cross-day drift) (ablation) | BRC · MNIST subset (200 images) | accuracy: 32 ± 3% on Day 2 and 28 ± 5% on Day 3 vs chance 9 ± 1% | Hurt |
| replace artificial reservoir with biological reservoir | reservoir type · Pointwise stimuli | accuracy: 82% ± 6% -> 98% ± 2% | Improved |
| replace artificial reservoir with biological reservoir | reservoir type · Oriented bars | accuracy: 98% ± 3% -> 92% ± 6% | Hurt |
| increase stimulus complexity from pointwise to overlapping oriented bars | BRC · Pointwise vs oriented bars | accuracy: 98% ± 2% -> 92% ± 6% | Hurt |
| stimulate on different days (Day 1 to Day 3), same-day training (ablation) | BRC · Clock-digit-like patterns | accuracy: 72% ± 7% -> 73% ± 8% -> 69% ± 7% | No clear effect |
| stimulate on different days (Day 1 to Day 3), same-day training (ablation) | BRC · MNIST subset (200 images) | accuracy: 37% ± 8% -> 34% ± 8% -> 33% ± 7% | No clear effect |
| use only digits with fewer stimulation electrodes (patterns 1 and 7) | BRC · Clock-digit-like patterns | accuracy | Improved |
Closest papers by meaning
- Bio-Vision-Inspired Spiking Neural Networks for Object Detection with Event Cameras (ICML 2026 · similarity 0.87)
- From deep learning to mechanistic understanding in neuroscience: the structure of retinal prediction (NeurIPS 2019 · similarity 0.87)
- Learning to play with spikes. Characterizing, predicting, and engineering unsupervised plasticity rules for spiking reservoir computing (NeurIPS 2026 · similarity 0.86)
- Recurrent neural network dynamical systems for biological vision (NeurIPS 2024 · similarity 0.86)
- Beyond Low-Pass Dynamics: Frequency-Selective Spiking Reservoirs with Resonant Neurons (NeurIPS 2026 · similarity 0.86)
- The Expressive Leaky Memory Neuron: an Efficient and Expressive Phenomenological Neuron Model Can Solve Long-Horizon Tasks. (ICLR 2024 · 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.
