RRR-Net: Reusing, Reducing, and Recycling a Deep Backbone Network - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

RRR-Net: Reusing, Reducing, and Recycling a Deep Backbone Network

Abstract

It has become mainstream in computer vision and other machine learning domains to reuse backbone networks pre-trained on large datasets as preprocessors. Typically, the last layer is replaced by a shallow learning machine of sorts; the newly-added classification head and (optionally) deeper layers are fine-tuned on a new task. Due to its strong performance and simplicity, a common pre-trained backbone network is ResNet152. However, ResNet152 is relatively large and induces inference latency. In many cases, a compact and efficient backbone with similar performance would be preferable over a larger, slower one. This paper investigates techniques to reuse a pre-trained backbone with the objective of creating a smaller and faster model. Starting from a large ResNet152 backbone pre-trained on ImageNet, we first reduce it from 51 blocks to 5 blocks, reducing its number of parameters and FLOPs by more than 6 times, without significant performance degradation. Then, we split the model after 3 blocks into several branches, while preserving the same number of parameters and FLOPs, to create an ensemble of sub-networks to improve performance. Our experiments on a large benchmark of $40$ image classification datasets from various domains suggest that our techniques match the performance (if not better) of ``classical backbone fine-tuning'' while achieving a smaller model size and faster inference speed.
Fichier principal
Vignette du fichier
main.pdf (720.81 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04224281 , version 1 (01-10-2023)

Identifiers

Cite

Haozhe Sun, Isabelle Guyon, Felix Mohr, Hedi Tabia. RRR-Net: Reusing, Reducing, and Recycling a Deep Backbone Network. IJCNN 2023 - International Joint Conference on Neural Networks, Jun 2023, Gold Coast, Australia. pp.1-9, ⟨10.1109/IJCNN54540.2023.10191770⟩. ⟨hal-04224281⟩
55 View
26 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More