Binarized Convolutional Neural Networks for Efficient Inference on GPUs
This enables efficient real-time inference on resource-constrained devices like embedded systems, though it is incremental as it builds on existing binarization techniques.
The paper tackles the computational expense of convolutional neural networks for image classification by implementing binarized networks on GPUs, achieving a maximum speedup of 7.4x with only a 4.4% accuracy loss compared to a floating-point reference.
Convolutional neural networks have recently achieved significant breakthroughs in various image classification tasks. However, they are computationally expensive,which can make their feasible mplementation on embedded and low-power devices difficult. In this paper convolutional neural network binarization is implemented on GPU-based platforms for real-time inference on resource constrained devices. In binarized networks, all weights and intermediate computations between layers are quantized to +1 and -1, allowing multiplications and additions to be replaced with bit-wise operations between 32-bit words. This representation completely eliminates the need for floating point multiplications and additions and decreases both the computational load and the memory footprint compared to a full-precision network implemented in floating point, making it well-suited for resource-constrained environments. We compare the performance of our implementation with an equivalent floating point implementation on one desktop and two embedded GPU platforms. Our implementation achieves a maximum speed up of 7. 4X with only 4.4% loss in accuracy compared to a reference implementation.