Convergence of Continual Learning in Homogeneous Deep Networks

arXiv:2606.30559 · cs.LG, math.NA, math.OC, stat.ML · Submitted 2026-06-29 · Read on arXiv

Matan Schliserman, Gon Buzaglo, Itay Evron, Daniel Soudry

cs.LG, math.NA, math.OC, stat.ML

Submitted: 2026-06-29

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Continual Learning — 45× in this paper · explained in 3 episodes
• Catastrophic Forgetting — 22× in this paper · explained in 6 episodes
• Regularization — 21× in this paper · explained in 3 episodes
• Editor — 7× in this paper · explained in 2 episodes
• Convergence Guarantees — 5× in this paper · explained in 3 episodes
• Hinge Loss — 5× in this paper · explained in 2 episodes
• Generalization — 4× in this paper · explained in 13 episodes
• Homogeneity — 3× in this paper · explained in 2 episodes
• Foundation Model — 2× in this paper · explained in 3 episodes
• Foundation Models — 2× in this paper · explained in 7 episodes
• Loss Function — 2× in this paper · explained in 2 episodes
• Projections Onto Convex Sets — 2× in this paper
• Activation Function — 1× in this paper · explained in 1 episode
• Adaptation — 1× in this paper · explained in 1 episode
• Core — 1× in this paper · explained in 2 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Feature Extraction — 1× in this paper · explained in 3 episodes
• Fisher Information — 1× in this paper · explained in 4 episodes
• Funded by the European Research Council — 1× in this paper
• Phase Transition — 1× in this paper · explained in 2 episodes
• Phase Transitions — 1× in this paper · explained in 2 episodes
• Weight Decay — 1× in this paper · explained in 2 episodes

Sources

• Re-evaluating Continual Learning Scenarios: A Categorization and Case for Strong Baselines
• Gradient Descent Maximizes the Margin of Homogeneous Neural Networks
• Order parameters and phase transitions of continual learning in deep neural networks
• On the Theory of Continual Learning with Gradient Descent for 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: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks