Research map / Architectures and efficiency / RNNs and normalization
Lamina-specific neuronal properties promote robust, stable signal propagation in feedforward networks
Dongqi Han, Erik De Schutter, Sungho Hong
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 |
|---|---|---|---|
| Use alternating integrator and differentiator layers instead of homogeneous layers | Deep feedforward network · Deep FFN, diverse input spike widths and counts | spike signal propagation stability | Improved |
| Use lamina-specific integrator PNs and differentiator LHNs instead of same-type PNs and LHNs | Drosophila AL network · Drosophila AL | information transfer | Improved |
| Use lamina-specific integrator PNs and differentiator LHNs instead of same-type PNs and LHNs | Drosophila AL network · Drosophila AL | input detection accuracy (d′) | Improved |
| Use lamina-specific integrator PNs and differentiator LHNs instead of same-type PNs and LHNs | Drosophila AL network · Drosophila AL | power amplification stability | Improved |
| Make PN and LHN properties homogeneous using differentiator neurons (ablation) | Drosophila AL network · Drosophila AL | input detection accuracy (d′) | Hurt |
| Make PN and LHN properties homogeneous using integrator neurons (ablation) | Drosophila AL network · Drosophila AL | information transfer | Hurt |
| Reverse the integrator/differentiator order across PN and LHN (ablation) | Drosophila AL network · Drosophila AL | information transfer | No clear effect |
| Reverse the integrator/differentiator order across PN and LHN (ablation) | Drosophila AL network · Drosophila AL | input detection accuracy (d′) | Hurt |
| Use homogeneous intermediate-type neurons instead of distinct cell types (ablation) | Drosophila AL network · Drosophila AL | information transfer | Hurt |
| Use homogeneous intermediate-type layers instead of alternating cell types (ablation) | Deep feedforward network · Deep FFN, diverse input spike widths and counts | spike signal propagation stability | Hurt |
| Add feedforward inhibition (ablation) | Deep feedforward network · Deep FFN | spike signal propagation | No clear effect |
| Vary integrator βw across the tested values (ablation) | Heterogeneous deep FFN · Deep FFN | signal propagation | No clear effect |
Closest papers by meaning
- Shaping the distribution of neural responses with interneurons in a recurrent circuit model (NeurIPS 2024 · similarity 0.84)
- Enhancing the Robustness of Spiking Neural Networks with Stochastic Gating Mechanisms (AAAI 2024 · similarity 0.83)
- Scaling Laws and Tradeoffs in Recurrent Networks of Expressive Neurons (NeurIPS 2026 · similarity 0.82)
- The Expressive Leaky Memory Neuron: an Efficient and Expressive Phenomenological Neuron Model Can Solve Long-Horizon Tasks. (ICLR 2024 · similarity 0.82)
- FEEL-SNN: Robust Spiking Neural Networks with Frequency Encoding and Evolutionary Leak Factor (NeurIPS 2024 · similarity 0.82)
- Self-Supervised Discovery of Neural Circuits in Spatially Patterned Neural Responses with Graph Neural Networks (NeurIPS 2025 · similarity 0.82)
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.
