English

Hydra Attention: Efficient Attention with Many Heads

Computer Vision and Pattern Recognition 2022-09-16 v1

Abstract

While transformers have begun to dominate many tasks in vision, applying them to large images is still computationally difficult. A large reason for this is that self-attention scales quadratically with the number of tokens, which in turn, scales quadratically with the image size. On larger images (e.g., 1080p), over 60% of the total computation in the network is spent solely on creating and applying attention matrices. We take a step toward solving this issue by introducing Hydra Attention, an extremely efficient attention operation for Vision Transformers (ViTs). Paradoxically, this efficiency comes from taking multi-head attention to its extreme: by using as many attention heads as there are features, Hydra Attention is computationally linear in both tokens and features with no hidden constants, making it significantly faster than standard self-attention in an off-the-shelf ViT-B/16 by a factor of the token count. Moreover, Hydra Attention retains high accuracy on ImageNet and, in some cases, actually improves it.

Keywords

Cite

@article{arxiv.2209.07484,
  title  = {Hydra Attention: Efficient Attention with Many Heads},
  author = {Daniel Bolya and Cheng-Yang Fu and Xiaoliang Dai and Peizhao Zhang and Judy Hoffman},
  journal= {arXiv preprint arXiv:2209.07484},
  year   = {2022}
}

Comments

Accepted CADL 2022 (ECCV Workshop)

R2 v1 2026-06-28T01:23:15.879Z