How to Train Your Metamorphic Deep Neural Network

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sommariva, Thomas, Calderara, Simone, Porrello, Angelo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914091791024128
author Sommariva, Thomas
Calderara, Simone
Porrello, Angelo
author_facet Sommariva, Thomas
Calderara, Simone
Porrello, Angelo
contents Neural Metamorphosis (NeuMeta) is a recent paradigm for generating neural networks of varying width and depth. Based on Implicit Neural Representation (INR), NeuMeta learns a continuous weight manifold, enabling the direct generation of compressed models, including those with configurations not seen during training. While promising, the original formulation of NeuMeta proves effective only for the final layers of the undelying model, limiting its broader applicability. In this work, we propose a training algorithm that extends the capabilities of NeuMeta to enable full-network metamorphosis with minimal accuracy degradation. Our approach follows a structured recipe comprising block-wise incremental training, INR initialization, and strategies for replacing batch normalization. The resulting metamorphic networks maintain competitive accuracy across a wide range of compression ratios, offering a scalable solution for adaptable and efficient deployment of deep models. The code is available at: https://github.com/TSommariva/HTTY_NeuMeta.
format Preprint
id arxiv_https___arxiv_org_abs_2505_05510
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle How to Train Your Metamorphic Deep Neural Network
Sommariva, Thomas
Calderara, Simone
Porrello, Angelo
Neural and Evolutionary Computing
Computer Vision and Pattern Recognition
Machine Learning
Neural Metamorphosis (NeuMeta) is a recent paradigm for generating neural networks of varying width and depth. Based on Implicit Neural Representation (INR), NeuMeta learns a continuous weight manifold, enabling the direct generation of compressed models, including those with configurations not seen during training. While promising, the original formulation of NeuMeta proves effective only for the final layers of the undelying model, limiting its broader applicability. In this work, we propose a training algorithm that extends the capabilities of NeuMeta to enable full-network metamorphosis with minimal accuracy degradation. Our approach follows a structured recipe comprising block-wise incremental training, INR initialization, and strategies for replacing batch normalization. The resulting metamorphic networks maintain competitive accuracy across a wide range of compression ratios, offering a scalable solution for adaptable and efficient deployment of deep models. The code is available at: https://github.com/TSommariva/HTTY_NeuMeta.
title How to Train Your Metamorphic Deep Neural Network
topic Neural and Evolutionary Computing
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2505.05510