We introduce Larq Compute Engine, the world's fastest Binarized Neural Network (BNN) inference engine, and use this framework to investigate several important questions about the efficiency of BNNs and to design a new state-of-the-art BNN architecture. LCE provides highly optimized implementations of binary operations and accelerates binary convolutions by 8.5 - 18.5x compared to their full-precision counterparts on Pixel 1 phones. LCE's integration with Larq and a sophisticated MLIR-based converter allow users to move smoothly from training to deployment. By extending TensorFlow and TensorFlow Lite, LCE supports models which combine binary and full-precision layers, and can be easily integrated into existing applications. Using LCE, we analyze the performance of existing BNN computer vision architectures and develop QuickNet, a simple, easy-to-reproduce BNN that outperforms existing binary networks in terms of latency and accuracy on ImageNet. Furthermore, we investigate the impact of full-precision shortcuts and the relationship between number of MACs and model latency. We are convinced that empirical performance should drive BNN architecture design and hope this work will facilitate others to design, benchmark and deploy binary models.
Co-Designing Binarized Transformer and Hardware Accelerator for Efficient End-to-End Edge Deployment
School of Electronic Science and Engineering, Nanjing University(南京大学电子科学与技术学院);Department of Computer Science and Engineering, Chinese University of Hong Kong(香港中文大学计算机科学与工程系);School of Integrated Circuits, Sun Yat-sen University(中山大学集成电路学院)
本文提出 BMT 二值化 Transformer、BAT 加速器及设计空间联合优化方法,在精度、延迟和鲁棒性间取得平衡,并显著提升边缘部署的吞吐量与能效。
MuBiNN: Multi-Level Binarized Recurrent Neural Network for EEG signal Classification
School of Electrical and Computer Engineering, University of Tehran(德黑兰大学电气与计算机工程学院);Division of Intelligent Future Technologies, Malardalen University(马拉达伦大学智能未来技术系)
An Automata-Theoretic Approach to Synthesizing Binarized Neural Networks
National University of Defense Technology(国防科技大学);ShanghaiTech University(上海科技大学);Institute of Software, Chinese Academy of Sciences(中国科学院软件研究所);University of Chinese Academy of Sciences(中国科学院大学);Automotive Software Innovation Center(汽车软件创新中心)
Binary Complex Neural Network Acceleration on FPGA
University of Connecticut(康涅狄格大学);Stevens Institute of Technology(史蒂文斯理工学院);University of Texas at San Antonio(德克萨斯大学圣安东尼奥分校);Pacific Northwest National Laboratory(太平洋西北国家实验室)
National Special Education Resource Center for Children with Autism, Zhejiang Normal University(浙江师范大学国家孤独症儿童特殊教育资源中心);Institute of Cyber-Systems and Control, Zhejiang University(浙江大学 cyber 系统与控制研究所);School of Computer Science and Technology, Zhejiang Normal University(浙江师范大学计算机科学与技术学院)
RBNN: Memory-Efficient Reconfigurable Deep Binary Neural Network with IP Protection for Internet of Things
Nanjing University of Science and Technology(南京理工大学);The University of Adelaide(阿德莱德大学);Data61, CSIRO(澳大利亚联邦科学与工业研究组织Data61);Harbin Institute of Technology(哈尔滨工业大学);Huazhong University of Science and Technology(华中科技大学)