generalization error
The difference between the model's performance on training data versus unseen test data. A low generalization error indicates the model can effectively apply learned knowledge to new inputs.
- A High-Dimensional Statistical Method for Optimizing Transfer Quantities in Multi-Source Transfer Learning
- Blindfolded Experts Generalize Better: Insights from Robotic Manipulation and Videogames
- Can DPO Learn Diverse Human Values? A Theoretical Scaling Law
- Conservative classifiers do consistently well with improving agents: characterizing statistical and online learning
- How Patterns Dictate Learnability in Sequential Data
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning
- On the Stability and Generalization of Meta-Learning: the Impact of Inner-Levels
- Tighter CMI-Based Generalization Bounds via Stochastic Projection and Quantization
- Tighter CMI-Based Generalization Bounds via Stochastic Projection and Quantization
- Towards Syn-to-Real IQA: A Novel Perspective on Reshaping Synthetic Data Distributions
- Unveiling the Power of Multiple Gossip Steps: A Stability-Based Generalization Analysis in Decentralized Training