Fast integer approximations in convolutional neural networks using layer-by-layer training-Reference-Cited by-同舟云学术

Fast integer approximations in convolutional neural networks using layer-by-layer training

Author:

Ilin Dmitry¹,Limonova Elena²,Arlazarov Vladimir²,Nikolaev Dmitry³

Affiliation:

1. Smart Engines Ltd. (Russian Federation)

2. Moscow Institute of Physics and Technology (Russian Federation)

3. Institute for Information Transmission Problems (Russian Federation)

Publisher

SPIE

Reference11 articles.

1. Viola-Jones based hybrid framework for real-time object detection in multispectral images;Kuznetsova,2015

2. Application of random ferns for non-planar object detection;Mastov,2015

3. Detection and tracking of vehicles based on the video-registration information;Kopenkov,2015

4. Improving neural network performance on SIMD architectures;Limonova,2015

5. Improving the speed of neural networks on CPUs;Vanhoucke;Deep Learning and Unsupervised Feature Learning Workshop,2011

Cited by 12 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

3. Optimizing Medical Image Classification Models for Edge Devices;Distributed Computing and Artificial Intelligence, Volume 1: 18th International Conference;2021-09-02

4. Fast Implementation of 4-bit Convolutional Neural Networks for Mobile Devices;2020 25th International Conference on Pattern Recognition (ICPR);2021-01-10