exercises-machine learning

XOR problem

build: 2016-07-16 solved: No

• XOR problem: [input, output1] pairs-- [(1, 0), 1], [(0, 1), 1], [(0, 0), 0], [(1, 1), 0]
• a two-layers network can approximate this XOR function, with 2 input layer nodes and 1 output node (sigmoid activation)

NAND problem

build: 2016-07-16 solved: No

• NAND problem: [input,, output] pairs-- [(1, 0), 1], [(0, 1), 1], [(0, 0), 1], [(1, 1), 0]
• may be a two-layers network can approximate this NAND function, with 2 input layer nodes and 1 output node (sigmoid activation)