Google Engineers 'Mutate' AI to Make It Evolve Systems Faster Than We Can Code Them
https://www.sciencealert.com/coders-mutate-ai-systems-to-make-them-evolve-faster-than-we-can-program-them
[2003.03384] AutoML-Zero: Evolving Machine Learning Algorithms From Scratch
https://arxiv.org/abs/2003.03384
Googleの自己進化するAI「AutoML Zero」 | スラド IT
https://it.srad.jp/story/20/04/17/1340217/
機械学習アルゴリズムを進化的検索で全自動生成する AutoML-Zero | AI-SCHOLAR.TECH
https://ai-scholar.tech/articles/automl/automl-zero-ai-391
3つの要点
✔ AutoML-Zeroは基本的な数学演算の操作だけで、完全な機械学習アルゴリズムを自動的に発見
✔ Back Propagationで学習するニューラルネットワークすら発見
✔ 人間の先入観に捉われない新たな解法の創造に期待が持てます。
AutoML(Automated Machine Learning)は、実際のデータに機械学習を適用するプロセスを自動化する技術です。
特定の問題に対して最適な機械学習モデルを自動構築する事も含みますが、これは主にニューラルネットワークに注目しており、
専門家が設計したレイヤーを構成要素として構築するため、モデルの探索空間が限られています。
AutoML-Zeroでは、基本的な数学演算を構成要素として使用するだけで、完全な機械学習アルゴリズムを自動的に発見します。
And it's fast too: the researchers reckon up to 10,000 possible algorithms can be searched through per second per processor (the more computer processors available for the task, the quicker it can work).
Eventually, this should see artificial intelligence systems become more widely used, and easier to access for programmers with no AI expertise.
It might even help us eradicate human bias from AI, because humans are barely involved.
Work to improve AutoML-Zero continues, with the hope that it'll eventually be able to spit out algorithms that mere human programmers would never have thought of.
Right now it's only capable of producing simple AI systems, but the researchers think the complexity can be scaled up rather rapidly.
"While most people were taking baby steps, [the researchers] took a giant leap into the unknown,"