Zach Furman
I’m trying to make AI safer by figuring out how it works, especially using tools from math and physics. Currently, I'm a PhD student at the University of Melbourne researching singular learning theory and the mathematical foundations of deep learning. I'm advised by Liam Hodgkinson, and collaborate closely with Daniel Murfet and Timaeus. I help run a weekly seminar on these topics every Thursday - come join! Before my PhD, I built rockets (embedded software, electrical, and aerospace engineering), and briefly did machine learning interpretability and condensed matter physics research. My undergraduate degree is in math and computer science from Boston University. Feel free to send me an email, schedule a 1-on-1 chat, or check out my writing and code.
Zach Furman Hi, I'm Zach! You can find my CV here . I’m trying to make AI safer by figuring out how it works, especially using tools from math and physics. Currently, I'm a PhD student at the University of Melbourne researching singular learning theory and the mathematical foundations of deep learning. I'm advised by Liam Hodgkinson , and collaborate closely with Daniel Murfet and Timaeus . I help run a weekly seminar on these topics every Thursday - come join! Before my PhD, I built rockets (embedded software, electrical, and aerospace engineering), and briefly did machine learning interpreta
related reading
- Vibe physics: The AI grad student \ Anthropicanthropic.com
- Study Guide — LessWronglesswrong.com
- Liu Ziyinmit.edu
- AI Safety: A Short FAQ for Mathematiciansalkjash.github.io
- Ziqian Zhongfjzzq2002.github.io
- fast.ai—Making neural nets uncool again – fast.aifast.ai
- Kushal Thamankushalthaman.github.io
- AI Safety for Mathematiciansmathforaisafety.org
- AI Safety for Mathematiciansmathforaisafety.org
- School 2.0 / Bio Pagebio.school2point0.com
- AI safety technical research | Career review | 80,000 Hours80000hours.org
- Singular Learning Theory Seminarzachfurman.com