Research map / Architectures and efficiency / RNNs and normalization
Credit Assignment Through Broadcasting a Global Error Vector
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 GEVB instead of BP | Nonnegative fully connected VNN · MNIST | test error (%): 0.14% maximum discrepancy in test error on MNIST | No clear effect |
| Use GEVB instead of BP | Nonnegative locally connected VNN · CIFAR-10 | test error (%): 0.59% maximum discrepancy in test error on MNIST and CIFAR-10 | No clear effect |
| Use GEVB instead of BP | Nonnegative convolutional VNN · MNIST and CIFAR-10 | test error (%): 1.33% and 2.89% discrepancies in test error on MNIST and CIFAR-10 | Hurt |
| Use GEVB in place of DFA | Conventional fully connected network · MNIST | test error (%) | Helped |
| Use GEVB in place of DFA | Convolutional network · CIFAR-10 | test error (%) | Helped |
| Use mixed-sign rather than nonnegative weights past the first layer (ablation) | GEVB-trained vectorized networks · MNIST and CIFAR-10 | test error (%) | Mixed |
| Impose vectorization or nonnegativity, or both (ablation) | BP-trained networks · MNIST and CIFAR-10 | test error (%): 8.69% | Hurt |
| Use GEVB rather than DFA in convolutional layers | Convolutional networks · CIFAR-10 | cluster quality | Helped |
Closest papers by meaning
- Error Broadcast and Decorrelation as a Potential Artificial and Natural Learning Mechanism (NeurIPS 2025 · similarity 0.87)
- Overcoming Rank Collapse in Feedback Alignment (NeurIPS 2026 · similarity 0.86)
- Forward Learning with Top-Down Feedback: Empirical and Analytical Characterization (ICLR 2024 · similarity 0.86)
- Counter-Current Learning: A Biologically Plausible Dual Network Approach for Deep Learning (NeurIPS 2024 · similarity 0.85)
- Feedback control guides credit assignment in recurrent neural networks (NeurIPS 2024 · similarity 0.85)
- Improving equilibrium propagation without weight symmetry through Jacobian homeostasis (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.