Researchers at MIT’s Computer Science & Artificial Intelligence Lab (CSAIL) have open-sourced Multiply-ADDitioN-lESS (MADDNESS), an algorithm that speeds up machine learning using approximate matrix multiplication (AMM). MADDNESS requires zero multiply-add operations and runs 10x faster than other approximate methods and 100x faster than exact multiplication.
The first author was gracious enough to answer questions about and detail the (many, many) caveats of their approach on HN. AIUI, this is not going to be a drop-in replacement for BLAS or anything anytime soon.