2021

Conditional Positional Encodings for Vision Transformers

Chu, Xiangxiang, Tian, Zhi, Zhang, Bo et al.

Understand

We propose a conditional positional encoding (CPE) scheme for vision Transformers.

Unlike previous fixed or learnable positional encodings, which are pre-defined and independent of input tokens, CPE is dynamically generated and conditioned on the local neighborhood of the input tokens.
As a result, CPE can easily generalize to the input sequences that are longer than what the model has ever seen during training.
Besides, CPE can keep the desired translation-invariance in the image classification task, resulting in improved performance.

Built on

Convolutional self-attention networks
Original
Baosong Yang, Longyue Wang, Derek Wong, Lidia S Chao, and Zhaopeng Tu · 1904
Earlier work this paper cites.
Xlnet: Generalized autoregressive pretraining for language understanding
Original
Zhilin Yang, Zihang Dai, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, and Quoc V Le · 1906
Earlier work this paper cites.
Imagenet: A large-scale hierarchical image database
Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei · 2009
Earlier work this paper cites.
Dropout: a simple way to prevent neural networks from overfitting
Nitish Srivastava, Geoffrey Hinton, Alex Krizhevsky, Ilya Sutskever, and Ruslan Salakhutdinov · 2014
Earlier work this paper cites.
Mxnet: A flexible and efficient machine learning library for heterogeneous distributed systems
Original
Tianqi Chen, Mu Li, Yutian Li, Min Lin, Naiyan Wang, Minjie Wang, Tianjun Xiao, Bing Xu, Chiyuan Zhang, and Zheng Zhang · 2015
Earlier work this paper cites.
Tensorflow: A system for large-scale machine learning
Martín Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al · 2016
Earlier work this paper cites.
Deep residual learning for image recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun · 2016
Earlier work this paper cites.
Deep networks with stochastic depth
Gao Huang, Yu Sun, Zhuang Liu, Daniel Sedra, and Kilian Q Weinberger · 2016
Earlier work this paper cites.
Rethinking the inception architecture for computer vision
Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jon Shlens, and Zbigniew Wojna · 2016
Earlier work this paper cites.
Focal loss for dense object detection
Tsung-Yi Lin, Priya Goyal, Ross Girshick, Kaiming He, and Piotr Dollár · 2017
Earlier work this paper cites.
Revisiting unreasonable effectiveness of data in deep learning era
Chen Sun, Abhinav Shrivastava, Saurabh Singh, and Abhinav Gupta · 2017
Earlier work this paper cites.
Attention is all you need
Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin · 2017
Earlier work this paper cites.
Scene parsing through ade20k dataset
Bolei Zhou, Hang Zhao, Xavier Puig, Sanja Fidler, Adela Barriuso, and Antonio Torralba · 2017
Earlier work this paper cites.
Music transformer
Cheng-Zhi Anna Huang, Ashish Vaswani, Jakob Uszkoreit, Noam Shazeer, Ian Simon, Curtis Hawthorne, Andrew M Dai, Matthew D Hoffman, Monica Dinculescu, and Douglas Eck · 2018
Earlier work this paper cites.
Image transformer
Niki Parmar, Ashish Vaswani, Jakob Uszkoreit, Lukasz Kaiser, Noam Shazeer, Alexander Ku, and Dustin Tran · 2018
Earlier work this paper cites.
Improving language understanding by generative pre-training
Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever · 2018
Earlier work this paper cites.
Self-attention with relative position representations
Peter Shaw, Jakob Uszkoreit, and Ashish Vaswani · 2018
Earlier work this paper cites.
Qanet: Combining local convolution with global self-attention for reading comprehension
Original
Adams Wei Yu, David Dohan, Minh-Thang Luong, Rui Zhao, Kai Chen, Mohammad Norouzi, and Quoc V Le · 2018
Earlier work this paper cites.
mixup: Beyond empirical risk minimization
Hongyi Zhang, Moustapha Cisse, Yann N. Dauphin, and David Lopez-Paz · 2018
Earlier work this paper cites.

Similar

Attention augmented convolutional networks
Irwan Bello, Barret Zoph, Ashish Vaswani, Jonathon Shlens, and Quoc V Le · 2019
Cited alongside, same era.
Bert: Pre-training of deep bidirectional transformers for language understanding
Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova · 2019
Cited alongside, same era.
Panoptic feature pyramid networks
Alexander Kirillov, Ross Girshick, Kaiming He, and Piotr Dollár · 2019
Cited alongside, same era.
Decoupled weight decay regularization
Ilya Loshchilov and Frank Hutter · 2019
Cited alongside, same era.
Transformers with convolutional context for asr
Original
Abdelrahman Mohamed, Dmytro Okhonko, and Luke Zettlemoyer · 2019
Cited alongside, same era.
Pytorch: An imperative style, high-performance deep learning library
Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Natalia Gimelshein, Luca Antiga, et al · 2019
Cited alongside, same era.
Language models are unsupervised multitask learners
Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever · 2019
Cited alongside, same era.
Stand-alone self-attention in vision models
Prajit Ramachandran, Niki Parmar, Ashish Vaswani, Irwan Bello, Anselm Levskaya, and Jon Shlens · 2019
Cited alongside, same era.
Efficientnet: Rethinking model scaling for convolutional neural networks
Mingxing Tan and Quoc Le · 2019
Cited alongside, same era.
Encoding word order in complex embeddings
Original
Benyou Wang, Donghao Zhao, Christina Lioma, Qiuchi Li, Peng Zhang, and Jakob Grue Simonsen · 2019
Cited alongside, same era.
Cutmix: Regularization strategy to train strong classifiers with localizable features
Sangdoo Yun, Dongyoon Han, Seong Joon Oh, Sanghyuk Chun, Junsuk Choe, and Youngjoon Yoo · 2019
Cited alongside, same era.
Are we done with imagenet?
Original
Lucas Beyer, Olivier J Hénaff, Alexander Kolesnikov, Xiaohua Zhai, and Aäron van den Oord · 2020
Cited alongside, same era.

Then

Augment your batch: Improving generalization through instance repetition
Elad Hoffer, Tal Ben-Nun, Itay Hubara, Niv Giladi, Torsten Hoefler, and Daniel Soudry · 2020
Later among the works it cites.
How much position information do convolutional neural networks encode?
Md Amirul Islam, Sen Jia, and Neil DB Bruce · 2020
Later among the works it cites.
Learning to encode position for transformer with continuous dynamical model
Xuanqing Liu, Hsiang-Fu Yu, Inderjit Dhillon, and Cho-Jui Hsieh · 2020
Later among the works it cites.
Designing network design spaces
Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, and Piotr Dollár · 2020
Later among the works it cites.
Training data-efficient image transformers & distillation through attention
Original
Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, and Hervé Jégou · 2020
Later among the works it cites.
End-to-end video instance segmentation with transformers
Original
Yuqing Wang, Zhaoliang Xu, Xinlong Wang, Chunhua Shen, Baoshan Cheng, Hao Shen, and Huaxia Xia · 2020
Later among the works it cites.
Rethinking semantic segmentation from a sequence-to-sequence perspective with transformers
Original
Sixiao Zheng, Jiachen Lu, Hengshuang Zhao, Xiatian Zhu, Zekun Luo, Yabiao Wang, Yanwei Fu, Jianfeng Feng, Tao Xiang, Philip HS Torr, et al · 2020
Later among the works it cites.
Random erasing data augmentation
Zhun Zhong, Liang Zheng, Guoliang Kang, Shaozi Li, and Yi Yang · 2020
Later among the works it cites.
Lambdanetworks: Modeling long-range interactions without attention
Irwan Bello · 2021
Closest in time.
Twins: Revisiting the design of spatial attention in vision transformers
Xiangxiang Chu, Zhi Tian, Yuqing Wang, Bo Zhang, Haibing Ren, Xiaolin Wei, Huaxia Xia, and Chunhua Shen · 2021
Closest in time.
Coatnet: Marrying convolution and attention for all data sizes
Zihang Dai, Hanxiao Liu, Quoc Le, and Mingxing Tan · 2021
Closest in time.
An image is worth 16x16 words: Transformers for image recognition at scale
Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, and Neil Houlsby · 2021
Closest in time.
Convit: Improving vision transformers with soft convolutional inductive biases
Stéphane d’Ascoli, Hugo Touvron, Matthew L Leavitt, Ari S Morcos, Giulio Biroli, and Levent Sagun · 2021
Closest in time.
Swin transformer: Hierarchical vision transformer using shifted windows
Original
Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo · 2021
Closest in time.
Roformer: Enhanced transformer with rotary position embedding
Original
Jianlin Su, Yu Lu, Shengfeng Pan, Bo Wen, and Yunfeng Liu · 2021
Closest in time.
Pyramid vision transformer: A versatile backbone for dense prediction without convolutions
Original
Wenhai Wang, Enze Xie, Xiang Li, Deng-Ping Fan, Kaitao Song, Ding Liang, Tong Lu, Ping Luo, and Ling Shao · 2021
Closest in time.
Cvt: Introducing convolutions to vision transformers
Original
Haiping Wu, Bin Xiao, Noel Codella, Mengchen Liu, Xiyang Dai, Lu Yuan, and Lei Zhang · 2021
Closest in time.
Deformable detr: Deformable transformers for end-to-end object detection
Xizhou Zhu, Weijie Su, Lewei Lu, Bin Li, Xiaogang Wang, and Jifeng Dai · 2021
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…