effective dimensionality
A measure of the intrinsic complexity of a model or dataset, indicating how many relevant features or dimensions are necessary to accurately represent or predict the underlying phenomena, potentially reducing overfitting.
- Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right One
- Slow Transition to Low-Dimensional Chaos in Heavy-Tailed Recurrent Neural Networks
- Stochastic Forward-Forward Learning through Representational Dimensionality Compression
- The Computational Advantage of Depth in Learning High-Dimensional Hierarchical Targets