Enregistré dans:
Détails bibliographiques
Auteurs principaux: Ramesh, Mahesh, Ramkumar, Aswinkumar
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:https://arxiv.org/abs/2312.01324
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916077857931264
author Ramesh, Mahesh
Ramkumar, Aswinkumar
author_facet Ramesh, Mahesh
Ramkumar, Aswinkumar
contents Recent studies have demonstrated the effectiveness of Gated Linear Units (GLU) in enhancing transformer models, particularly in Large Language Models (LLMs). Additionally, utilizing a parallel configuration within each Transformer block rather than the conventional serialized method has been revealed to accelerate the training of LLMs without significantly impacting performance. However, when the MLP and attention block were run in parallel for the image classification task, we observed a noticeable decline in performance. We propose a novel transformer variant that integrates non-linearity within the attention block to tackle this problem. We implemented the GLU-based activation function on the Value tensor, and this new technique surpasses the current state-of-the-art S/16 variant of Vision Transformers by 0.6% on the ImageNet-1K dataset while utilizing fewer parameters. It also supersedes the B/16 variant while using only half the parameters. Furthermore, we provide results with the GELU activation function variant to confirm our assertions. Lastly, we showcase that the MABViT variants exhibit greater potential when utilized in deep transformers compared to the standard architecture.
format Preprint
id arxiv_https___arxiv_org_abs_2312_01324
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MABViT -- Modified Attention Block Enhances Vision Transformers
Ramesh, Mahesh
Ramkumar, Aswinkumar
Computer Vision and Pattern Recognition
Machine Learning
Recent studies have demonstrated the effectiveness of Gated Linear Units (GLU) in enhancing transformer models, particularly in Large Language Models (LLMs). Additionally, utilizing a parallel configuration within each Transformer block rather than the conventional serialized method has been revealed to accelerate the training of LLMs without significantly impacting performance. However, when the MLP and attention block were run in parallel for the image classification task, we observed a noticeable decline in performance. We propose a novel transformer variant that integrates non-linearity within the attention block to tackle this problem. We implemented the GLU-based activation function on the Value tensor, and this new technique surpasses the current state-of-the-art S/16 variant of Vision Transformers by 0.6% on the ImageNet-1K dataset while utilizing fewer parameters. It also supersedes the B/16 variant while using only half the parameters. Furthermore, we provide results with the GELU activation function variant to confirm our assertions. Lastly, we showcase that the MABViT variants exhibit greater potential when utilized in deep transformers compared to the standard architecture.
title MABViT -- Modified Attention Block Enhances Vision Transformers
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2312.01324