Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions
Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions
Neural networks rely on convolutions to aggregate spatial information. However, spatial convolutions are expensive in terms of model size and computation, both of which grow quadratically with respect to kernel size. In this paper, we present a parameter-free, FLOP-free "shift" operation as an alternative to spatial convolutions. We fuse shifts …