Research map / Architectures and efficiency / RNNs and normalization
Attention-Gated Brain Propagation: How the brain can implement reward-based error backpropagation
Isabella Pozzi, Sander Bohte, Pieter Roelfsema
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 BrainProp reward-based learning instead of error backprop (Error-BP) | BrainProp · Tiny ImageNet deep conv | accuracy (%): 47.35 -> 47.50 | No clear effect |
| Use BrainProp instead of Error-BP | BrainProp · CIFAR100 deep conv | accuracy (%): 63.39 -> 59.58 | Hurt |
| Use BrainProp instead of Error-BP | BrainProp · CIFAR10 deep conv | accuracy (%): 88.48 -> 88.88 | No clear effect |
| Use BrainProp instead of Error-BP | BrainProp · MNIST conv | accuracy (%): 99.39 -> 99.31 | No clear effect |
| Use BrainProp instead of Error-BP | BrainProp · MNIST dense | accuracy (%): 98.54 -> 98.68 | No clear effect |
| Use BrainProp instead of Error-BP | BrainProp · CIFAR10 conv | accuracy (%): 70.69 -> 69.30 | Hurt |
| Use BrainProp instead of Error-BP | BrainProp · CIFAR100 conv | accuracy (%): 36.56 -> 34.07 | Hurt |
| Use locally connected layers instead of shared-weight convolution (ablation) | BrainProp · CIFAR10 | accuracy (%): 69.30 -> 64.92 | Hurt |
| Use BrainProp instead of Error-BP | BrainProp · training epochs, CIFAR100 deep conv | epochs to converge: 101 -> 218 | Hurt |
Closest papers by meaning
- Can the Brain Do Backpropagation? --- Exact Implementation of Backpropagation in Predictive Coding Networks (NeurIPS 2020 · similarity 0.89)
- Assessing the Scalability of Biologically-Motivated Deep Learning Algorithms and Architectures (NeurIPS 2018 · similarity 0.89)
- Feedback control guides credit assignment in recurrent neural networks (NeurIPS 2024 · similarity 0.87)
- Efficient Target Propagation by Deriving Analytical Solution (AAAI 2024 · similarity 0.87)
- Stay Hungry, Keep Learning: Sustainable Plasticity for Deep Reinforcement Learning (ICML 2025 · similarity 0.86)
- Two Tales of Single-Phase Contrastive Hebbian Learning (ICML 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.
