Jan Leike on how to become a machine learning alignment researcher

Abstract

The article provides a practical guide for aspiring machine learning alignment researchers, drawing on the experience of Jan Leike, a Research Scientist at DeepMind. The article emphasizes the need for a strong foundation in mathematics and computer science, particularly linear algebra, calculus, algorithms, and statistics. It also recommends gaining practical research experience, potentially through publishing a paper before applying for a PhD. The article discusses various ways to contribute to machine learning alignment, such as pursuing a PhD in machine learning, engaging in policy work, or supporting research through philanthropy. The article highlights the significant talent gap in the field of technical AI safety and encourages individuals with a strong machine learning background to consider joining the effort. – AI-generated abstract