Conditioned Direct Feedback Alignment via Activity and Error Geometry
cs.LG, cs.NE, q-bio.NC
Submitted: 2026-07-20
Updated: 2026-09-27
Code: https://github.com/KempnerInstitute/conditioned-dfa
Terminology
Sources
- Correlations Are Ruining Your Gradient Descent
- Invariant Risk Minimization
- Overcoming Rank Collapse in Feedback Alignment
- Node Perturbation Can Effectively Train Multi-Layer Neural Networks
- Efficient Deep Learning with Decorrelated Backpropagation
- The Forward-Forward Algorithm: Some Preliminary Investigations
- Credit Assignment via Neural Manifold Noise Correlation
- Feedback alignment in deep convolutional networks
- How Does Preconditioning Guide Feature Learning in Deep Neural Networks?
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks