Local Binary Pattern Networks
This work addresses the need for efficient deep learning architectures, particularly for small footprint devices and hardware accelerators, though it appears incremental as it builds on existing binarization strategies.
The paper tackles the problem of memory and computation efficiency in deep learning by proposing Local Binary Pattern Networks (LBPNet), which uses local binary comparisons and random projections to replace conventional convolutions, achieving promising results on standard benchmarks while improving efficiency for small devices and hardware accelerators.
Memory and computation efficient deep learning architec- tures are crucial to continued proliferation of machine learning capabili- ties to new platforms and systems. Binarization of operations in convo- lutional neural networks has shown promising results in reducing model size and computing efficiency. In this paper, we tackle the problem us- ing a strategy different from the existing literature by proposing local binary pattern networks or LBPNet, that is able to learn and perform binary operations in an end-to-end fashion. LBPNet1 uses local binary comparisons and random projection in place of conventional convolu- tion (or approximation of convolution) operations. These operations can be implemented efficiently on different platforms including direct hard- ware implementation. We applied LBPNet and its variants on standard benchmarks. The results are promising across benchmarks while provid- ing an important means to improve memory and speed efficiency that is particularly suited for small footprint devices and hardware accelerators.