1 paper
Sean Wu, Jordan Hong, Keyu Bai +1
Due to their weak inductive bias, Multi-Layer Perceptrons (MLPs) have subpar performance at low-compute levels compared to standard architectures such as convolution-based networks…