Saved in:
Bibliographic Details
Main Authors: Sun, Yuwei, Ochiai, Hideya, Wu, Zhirong, Lin, Stephen, Kanai, Ryota
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2309.12862
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915190192209920
author Sun, Yuwei
Ochiai, Hideya
Wu, Zhirong
Lin, Stephen
Kanai, Ryota
author_facet Sun, Yuwei
Ochiai, Hideya
Wu, Zhirong
Lin, Stephen
Kanai, Ryota
contents Emerging from the pairwise attention in conventional Transformers, there is a growing interest in sparse attention mechanisms that align more closely with localized, contextual learning in the biological brain. Existing studies such as the Coordination method employ iterative cross-attention mechanisms with a bottleneck to enable the sparse association of inputs. However, these methods are parameter inefficient and fail in more complex relational reasoning tasks. To this end, we propose Associative Transformer (AiT) to enhance the association among sparsely attended input tokens, improving parameter efficiency and performance in various vision tasks such as classification and relational reasoning. AiT leverages a learnable explicit memory comprising specialized priors that guide bottleneck attentions to facilitate the extraction of diverse localized tokens. Moreover, AiT employs an associative memory-based token reconstruction using a Hopfield energy function. The extensive empirical experiments demonstrate that AiT requires significantly fewer parameters and attention layers outperforming a broad range of sparse Transformer models. Additionally, AiT outperforms the SOTA sparse Transformer models including the Coordination method on the Sort-of-CLEVR dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2309_12862
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Associative Transformer
Sun, Yuwei
Ochiai, Hideya
Wu, Zhirong
Lin, Stephen
Kanai, Ryota
Machine Learning
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
Emerging from the pairwise attention in conventional Transformers, there is a growing interest in sparse attention mechanisms that align more closely with localized, contextual learning in the biological brain. Existing studies such as the Coordination method employ iterative cross-attention mechanisms with a bottleneck to enable the sparse association of inputs. However, these methods are parameter inefficient and fail in more complex relational reasoning tasks. To this end, we propose Associative Transformer (AiT) to enhance the association among sparsely attended input tokens, improving parameter efficiency and performance in various vision tasks such as classification and relational reasoning. AiT leverages a learnable explicit memory comprising specialized priors that guide bottleneck attentions to facilitate the extraction of diverse localized tokens. Moreover, AiT employs an associative memory-based token reconstruction using a Hopfield energy function. The extensive empirical experiments demonstrate that AiT requires significantly fewer parameters and attention layers outperforming a broad range of sparse Transformer models. Additionally, AiT outperforms the SOTA sparse Transformer models including the Coordination method on the Sort-of-CLEVR dataset.
title Associative Transformer
topic Machine Learning
Computer Vision and Pattern Recognition
Neural and Evolutionary Computing
url https://arxiv.org/abs/2309.12862