paper
FFR: Forward-Forward Learning for Regression
paperactiveprovisional
ffr-forward-forward-learning-for-regression-925d93e3·1 events·first seen 13d agoAliases: FFR: Forward-Forward Learning for Regression
Co-occurring entities
More like this (12)
Forward-Forward AlgorithmExpRL: Exploratory RL for LLM Mid-TrainingRange Penalization: Theoretical Insights with Applications in Federated LearningFederated LearningFlashbackCL: Mitigating Temporal Forgetting in Federated LearningFeedforward Neural NetworksRLIF (Reinforcement Learning from Internal Feedback)KL-regularized RLLearning from the Self-future: On-policy Self-distillation for dLLMsEntropy-Regularized Reinforcement LearningUnsupervised Continual Clustering via Forward-Backward Knowledge DistillationTailLoR: Protecting Principal Components in Parameter-Efficient Continual Learning
Recent events (1)
FFR extends Forward-Forward algorithm to regression tasks with 73% memory reduction
A new arXiv preprint introduces FFR (Forward-Forward for Regression), the first framework to adapt Hinton's Forward-Forward algorithm—a biologically plausible, backpropagation-free training method—to regression problems. FFR introduces an ordinal competitive goodness function, a stratified ladder architecture, and hierarchical prediction with uncertainty estimation to handle continuous target spaces. Across five real-world regression benchmarks, FFR recovers 98.6% of backpropagation accuracy while reducing peak training memory to 27% of BP's at depth 8 and 8% at depth 32, with per-iteration time around 72% of BP's.