myffical·3개월 전·discussSome DeepMind researchers used mechanistic interpretability techniques to find concepts in AlphaZero and teach them to human chess Grandmasters: https://www.pnas.org/doi/10.1073/pnas.2406675122
myffical·5개월 전·discussAnother appropriate analogy would be be HeLa / Henrietta Lacks: https://en.wikipedia.org/wiki/HeLa