BMXNet: An Open-Source Binary Neural Network Implementation Based on MXNet
This work provides a practical tool for deploying efficient deep learning models on resource-constrained devices, but it is incremental as it implements existing BNN methods in a new framework.
The authors tackled the problem of enabling state-of-the-art deep learning on low-power devices by developing BMXNet, an open-source library for Binary Neural Networks (BNNs) that reduces memory and energy usage, with extensive experiments validating its efficiency and effectiveness.
Binary Neural Networks (BNNs) can drastically reduce memory size and accesses by applying bit-wise operations instead of standard arithmetic operations. Therefore it could significantly improve the efficiency and lower the energy consumption at runtime, which enables the application of state-of-the-art deep learning models on low power devices. BMXNet is an open-source BNN library based on MXNet, which supports both XNOR-Networks and Quantized Neural Networks. The developed BNN layers can be seamlessly applied with other standard library components and work in both GPU and CPU mode. BMXNet is maintained and developed by the multimedia research group at Hasso Plattner Institute and released under Apache license. Extensive experiments validate the efficiency and effectiveness of our implementation. The BMXNet library, several sample projects, and a collection of pre-trained binary deep models are available for download at https://github.com/hpi-xnor