90/30 ML Reading Group: Retrospective - by Lydia Nottingham
lydianottingham.substack.com · 1,266 words · saved by 1 readers
This corner of the SF AI scene is wonderful
To skip to the technical content, head to Section 5. For seven weeks, I ran an ML reading group in San Francisco. Here are 13 things I learned. Cross-context groups can work well In the first week, I was nervous about how more experienced attendees / people who work in ML full-time would find the group. I wanted to keep it open to all backgrounds, & I was on the newer end. I know it’s hard to have great discussions when participants lack context. To this end, I set up a ‘no dumb questions’ room and a ‘prior knowledge assumed’ room. This worked great—for the ‘no dumb questions’ room! No…
related reading
- James Mulhollandjmulholland.com
- School, Homenotes.ekzhang.com
- The Rationality Community Sucksuli.rocks
- Gathering Structuresmaggieappleton.com
- How I became a machine learning practitionerblog.gregbrockman.com
- Alignment Research Field Guide — AI Alignment Forumalignmentforum.org
- Humans are not automatically strategic — LessWronglesswrong.com
- Group chats rule the worldsriramk.com
- Alignment Research Field Guide — LessWronglesswrong.com
- CompCogSci Reading Group - Homecompucogsci.github.io
- MATS 9 Retrospective & Advice — LessWronglesswrong.com
- How to Run a Paper Club (also: LIVE at NeurIPS 2024!)latent.space