Algorithmic Simplification of Neural Networks with Mosaic-of-Motifs

ArXi:2602.14896v2 Announce Type: replace Large-scale deep learning models are well-suited for compression. Across a variety of tasks, methods like pruning, quantization, and knowledge distillation have been used to achieve massive reductions in model parameters with only marginal performance drops. This raises the central question: *Why are deep neural networks suited for compression?* In this work, we take up the perspective of algorithmic complexity to explain this behavior.