On: An In-Depth Look at Google's Tensor Processing Uni...

[yorwba]

You can get the full 32-bit result in only three multiplications using https://en.wikipedia.org/wiki/Karatsuba_algorithm , but it requires more additions.

[stephencanon → yorwba]

Yeah, in practice Karatsuba isn't a performant option for small operands unless your multiplier is catastrophically slow. (And it still doesn't get you to two multiplications.)