Papers
arxiv:2212.14215

Local Learning on Transformers via Feature Reconstruction

Published on Dec 29, 2022
Authors:
,

Abstract

A new local learning method for transformers reconstructs input features instead of entire images, reducing memory usage and improving performance compared to InfoPro and end-to-end training.

AI-generated summary

Transformers are becoming increasingly popular due to their superior performance over conventional convolutional neural networks(CNNs). However, transformers usually require a much larger amount of memory to train than CNNs, which prevents their application in many low resource settings. Local learning, which divides the network into several distinct modules and trains them individually, is a promising alternative to the end-to-end (E2E) training approach to reduce the amount of memory for training and to increase parallelism. This paper is the first to apply Local Learning on transformers for this purpose. The standard CNN-based local learning method, InfoPro [32], reconstructs the input images for each module in a CNN. However, reconstructing the entire image does not generalize well. In this paper, we propose a new mechanism for each local module, where instead of reconstructing the entire image, we reconstruct its input features, generated from previous modules. We evaluate our approach on 4 commonly used datasets and 3 commonly used decoder structures on Swin-Tiny. The experiments show that our approach outperforms InfoPro-Transformer, the InfoPro with Transfomer backbone we introduced, by at up to 0.58% on CIFAR-10, CIFAR-100, STL-10 and SVHN datasets, while using up to 12% less memory. Compared to the E2E approach, we require 36% less GPU memory when the network is divided into 2 modules and 45% less GPU memory when the network is divided into 4 modules.

Community

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2212.14215 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2212.14215 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2212.14215 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.