Characterizing Learning in Deep Neural Networks using Tractable Algorithmic Complexity Analysis
cs.LG
Submitted: 2026-05-15
Updated: 2026-11-09
Code: https://github.com/saintslab/QuBD
Terminology
Sources
- Emergence of Invariance and Disentanglement in Deep Representations
- Algorithmic Simplification of Neural Networks with Mosaic-of-Motifs
- A Survey of Quantization Methods for Efficient Neural Network Inference
- Compressing Deep Convolutional Networks using Vector Quantization
- SETOL: A Semi-Empirical Theory of (Deep) Learning
- A White Paper on Neural Network Quantization
- Binarized Neural Networks Converge Toward Algorithmic Simplicity: Empirical Support for the Learning-as-Compression Hypothesis
- Opening the Black Box of Deep Neural Networks via Information
- The information bottleneck method
- Deep Learning is Not So Mysterious or Different
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- Two-Dimensional Kolmogorov Complexity and Validation of the Coding Theorem Method by Compressibility
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis