Cumulative-Goodness Free-Riding in Forward-Forward Networks: Real, Repairable, but Not Accuracy-Dominant
cs.LG, cs.AI
Submitted: 2026-05-07
Updated: 2026-09-28
Comments: Accepted at NeurIPS 2026 (poster). Camera-ready version: 65 pages (10 pages main text), 24 figures, 45 tables. Code: https://github.com/amirhossein-yousefi/ff-free-riding ; checkpoints: https://huggingface.co/Amirhossein75/ff-free-riding
Code: https://github.com/amirhossein-yousefi/ff-free-riding
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Contrastive Forward-Forward: A Training Algorithm of Vision Transformer
- The Trifecta: Three simple techniques for training deeper Forward-Forward networks
- Extending the Forward Forward Algorithm
- Adaptive Spatial Goodness Encoding: Advancing and Scaling Forward-Forward Learning Without Backpropagation
- The Forward-Forward Algorithm: Some Preliminary Investigations
- Scalable Forward-Forward Algorithm
- Layer Collaboration in the Forward-Forward Algorithm
- Convolutional Channel-wise Competitive Learning for the Forward-Forward Algorithm
- Hyperspherical Forward-Forward with Prototypical Representations
- GLU Variants Improve Transformer
- Distance-Forward Learning: Enhancing the Forward-Forward Algorithm Towards High-Performance On-Chip Learning
- Cyclic Neural Network
- ST-MoE: Designing Stable and Transferable Sparse Expert Models
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: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks