Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformers

计算机科学 人工智能 安全性令牌 频域 算法 理论计算机科学 计算机视觉 计算机安全
作者
John Guibas,Morteza Mardani,Zongyi Li,Andrew Tao,Anima Anandkumar,Bryan Catanzaro
出处
期刊:Cornell University - arXiv 被引量:52
标识
DOI:10.48550/arxiv.2111.13587
摘要

Vision transformers have delivered tremendous success in representation learning. This is primarily due to effective token mixing through self attention. However, this scales quadratically with the number of pixels, which becomes infeasible for high-resolution inputs. To cope with this challenge, we propose Adaptive Fourier Neural Operator (AFNO) as an efficient token mixer that learns to mix in the Fourier domain. AFNO is based on a principled foundation of operator learning which allows us to frame token mixing as a continuous global convolution without any dependence on the input resolution. This principle was previously used to design FNO, which solves global convolution efficiently in the Fourier domain and has shown promise in learning challenging PDEs. To handle challenges in visual representation learning such as discontinuities in images and high resolution inputs, we propose principled architectural modifications to FNO which results in memory and computational efficiency. This includes imposing a block-diagonal structure on the channel mixing weights, adaptively sharing weights across tokens, and sparsifying the frequency modes via soft-thresholding and shrinkage. The resulting model is highly parallel with a quasi-linear complexity and has linear memory in the sequence size. AFNO outperforms self-attention mechanisms for few-shot segmentation in terms of both efficiency and accuracy. For Cityscapes segmentation with the Segformer-B3 backbone, AFNO can handle a sequence size of 65k and outperforms other efficient self-attention mechanisms.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
辞清完成签到 ,获得积分10
1秒前
dahuahau完成签到,获得积分10
1秒前
dcdedgbvr完成签到,获得积分20
1秒前
orixero应助顺利的乌冬面采纳,获得30
1秒前
耐斯糖完成签到 ,获得积分10
1秒前
山东人在南京完成签到 ,获得积分10
2秒前
柒柒发布了新的文献求助10
2秒前
酷波er应助Ted采纳,获得10
2秒前
啊啊啊完成签到,获得积分10
3秒前
Somnolence咩完成签到,获得积分10
3秒前
3秒前
虚幻的凤完成签到,获得积分10
4秒前
4秒前
开朗的汉堡完成签到,获得积分10
4秒前
大胆飞荷完成签到,获得积分10
4秒前
123完成签到,获得积分10
5秒前
可可完成签到,获得积分10
5秒前
万能图书馆应助忌辛辣采纳,获得10
6秒前
陶醉世德完成签到,获得积分10
6秒前
殷勤的紫槐完成签到,获得积分10
6秒前
7秒前
7秒前
7秒前
7秒前
lichunlei完成签到,获得积分10
8秒前
8秒前
欧阳振完成签到,获得积分10
9秒前
bigpluto完成签到,获得积分10
9秒前
白石溪完成签到,获得积分10
9秒前
lzh发布了新的文献求助10
10秒前
Aloha完成签到 ,获得积分10
10秒前
伶俐芷珊发布了新的文献求助10
10秒前
彩色蓉完成签到,获得积分10
11秒前
shisui完成签到,获得积分10
11秒前
酷酷的雪糕完成签到,获得积分10
11秒前
深情安青应助天舞英姿采纳,获得10
11秒前
NiNi完成签到,获得积分20
11秒前
白桃战士完成签到,获得积分10
11秒前
Z2WWS32发布了新的文献求助10
12秒前
精明的寒天完成签到,获得积分10
13秒前
高分求助中
A new approach to the extrapolation of accelerated life test data 1000
Cognitive Neuroscience: The Biology of the Mind 1000
Technical Brochure TB 814: LPIT applications in HV gas insulated switchgear 1000
Immigrant Incorporation in East Asian Democracies 600
Nucleophilic substitution in azasydnone-modified dinitroanisoles 500
不知道标题是什么 500
A Preliminary Study on Correlation Between Independent Components of Facial Thermal Images and Subjective Assessment of Chronic Stress 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 内科学 纳米技术 计算机科学 化学工程 复合材料 遗传学 基因 物理化学 催化作用 冶金 细胞生物学 免疫学
热门帖子
关注 科研通微信公众号,转发送积分 3968719
求助须知:如何正确求助?哪些是违规求助? 3513608
关于积分的说明 11168681
捐赠科研通 3248960
什么是DOI,文献DOI怎么找? 1794573
邀请新用户注册赠送积分活动 875194
科研通“疑难数据库(出版商)”最低求助积分说明 804716