Bias-variance decompositions: the exclusive privilege of Bregman divergences

arXiv:2501.18581 · cs.LG · Submitted 2025-01-30 · Read on arXiv

cs.LG

Submitted: 2025-01-30

Updated: 2026-01-25

Terminology

• Loss Function — 75× in this paper · explained in 2 episodes
• Generalization — 8× in this paper · explained in 13 episodes
• Additivity — 1× in this paper · explained in 1 episode
• Clustering — 1× in this paper · explained in 2 episodes
• Ensemble Learning — 1× in this paper · explained in 2 episodes

Sources

• A Modern Take on the Bias-Variance Tradeoff in Neural Networks
• Generalized Bregman and Jensen divergences which include some f-divergences
• A Generalized Bias-Variance Decomposition for Bregman Divergences

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