Research map / Optimization / Implicit bias and generalization
Convergence and Alignment of Gradient Descent with Random Backpropagation Weights
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 |
|---|---|---|---|
| add l2 regularization (lambda > 0) to feedback alignment (ablation) | two-layer networks (linear, ReLU, Tanh) · synthetic data, wide hidden layers | alignment of forward and backward weights (cosine) | Helped |
| add dropout as regularization instead of l2 (ablation) | feedback alignment two-layer networks · synthetic data, large hidden width | alignment | Hurt |
| increase l2 regularization lambda from 0 to 0.1 and 0.3 (ablation) | feedback alignment 784-1000-10 network · MNIST | alignment and test accuracy | Helped |
| increase l2 regularization lambda (ablation) | feedback alignment two-layer networks · synthetic data | training loss convergence speed | Hurt |
Closest papers by meaning
- Overcoming Rank Collapse in Feedback Alignment (NeurIPS 2026 · similarity 0.89)
- Forward Learning with Top-Down Feedback: Empirical and Analytical Characterization (ICLR 2024 · similarity 0.89)
- Pretraining with Random Noise for Fast and Robust Learning without Weight Transport (NeurIPS 2024 · similarity 0.89)
- Implicit Regularization in Feedback Alignment Learning Mechanisms for Neural Networks (ICML 2024 · similarity 0.88)
- Improving Deep Learning Speed and Performance Through Synaptic Neural Balance (AAAI 2025 · similarity 0.88)
- Efficient Target Propagation by Deriving Analytical Solution (AAAI 2024 · similarity 0.88)
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.