BCNN: A Binary CNN with All Matrix Ops Quantized to 1 Bit Precision
This paper describes a CNN where all CNN style 2D convolution operations that lower to matrix matrix multiplication are fully binary. The network is derived from a common building block structure that is consistent with a constructive proof outline showing that binary neural networks are universal function approximators. 68.96 with a 2 step training procedure and implementation strategies optimized for binary operands are provided.
READ FULL TEXT