Research map / Architectures and efficiency / RNNs and normalization
Single-phase deep learning in cortico-cortical networks
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 |
|---|---|---|---|
| use random fixed feedback weights instead of W-Y symmetric feedback (ablation) | BurstCCN · CIFAR-10 | test error (%): 22.92 -> 38.99 | Hurt |
| add STD Q feedback connections with homeostatic plasticity (BurstCCN vs EDN) | EDN · MNIST 5-layer | test error (%): 10.65 -> 1.84 | Helped |
| use single-phase learning instead of two-phase learning (ablation) | Burstprop · spiking XOR | task learned | Hurt |
| use single-phase learning instead of two-phase learning (ablation) | BurstCCN · spiking XOR | task learned | No clear effect |
| increase network depth (ablation) | EDN · MNIST | test error | Hurt |
| increase network depth (ablation) | BurstCCN · MNIST | test error | No clear effect |
| learn Q weights to cancel apical potential (ablation) | BurstCCN · 5-layer alignment to backprop | alignment angle | Helped |
| fix Q-Y symmetric (idealised) instead of learning Q (ablation) | BurstCCN · MNIST | alignment to backprop | Helped |
Closest papers by meaning
- Dendritic cortical microcircuits approximate the backpropagation algorithm (NeurIPS 2018 · similarity 0.91)
- Feedback control guides credit assignment in recurrent neural networks (NeurIPS 2024 · similarity 0.86)
- Scaling Laws and Tradeoffs in Recurrent Networks of Expressive 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)
- Emergent mechanisms for long timescales depend on training curriculum and affect performance in memory tasks (ICLR 2024 · similarity 0.85)
- Partial observation can induce mechanistic mismatches in data-constrained models of neural dynamics (NeurIPS 2024 · similarity 0.84)
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.