What is the easiest/funnest way to build up a comprehensive understanding of AI and AI Safety?

post by Jordan Arel · 2024-04-30T18:41:53.017Z · LW · GW · No comments

This is a question post.

Contents

  Answers
    4 Drake Morrison
None
No comments

I have spent around 100–200 hours listening to AI safety audiobooks, AI Safety Fundamentals course, Rob Miles YouTube, The Sequences, various bits and pieces of a bunch of YouTube AI channels and podcasts, as well as some time thinking through the basic case for X-risk.

When I look at certain heavy academic stuff or try to consume more technical content I sometimes get pretty lost. I am wondering how I can best build up my understanding of the basic technical details and terminology of AI and a broad overview of AI safety, such that I don’t get lost and it isn’t too grueling or over my head.

For context, fun doesn’t necessarily have to mean entertaining, I would find it fun to read a textbook that I can understand and that gradually builds my knowledge;

And easy doesn’t have to mean super basic, it probably needs to start with basics (I know up to pre-calculus but am unusually good at learning new math, I know relatively little about computer science) but then I would like to gradually build up to have a relatively deep understanding of whatever mathematics and technical details I need to really understand the problems at hand.

But easy and fun could also mean really informative podcasts or fiction that actually provides really useful insights, or a YouTube channel that explains the basics really effectively.

I guess the basic idea is I want to focus hard-core on AI, but I don’t want to burn out and want to ease into it in a way that makes me excited about it and enjoy it as much as possible, and I am wondering if anyone has experienced such content themselves, if so I would love to hear about it!

Thanks in advance!

Answers

answer by Drake Morrison · 2024-04-30T19:40:32.446Z · LW(p) · GW(p)

If you can code, build a small AI with the fast.ai course. This will (hopefully) be fun while also showing you particular holes in your knowledge to improve, rather than a vague feeling of "learn more". 

If you want to follow along with more technical papers, you need to know the math of machine learning: linear algebra, multivariable calculus, and probability theory. For Agent Foundations work, you'll need more logic and set theory type stuff. 

MIRI has some recommendations for textbooks here. There's also the Study Guide [LW · GW] and this sequence on leveling up [? · GW].

3blue1brown's Youtube has good videos for a lot of this, if that's the medium you like. 

If you like non-standard fiction, some people like Project Lawful.
 

At the end of the day, it's not a super well-defined field that has clear on-ramps into the deeper ends. You just gotta start somewhere, and follow your curiosity. Have fun!

↑ comment by metachirality · 2024-04-30T20:05:11.016Z · LW(p) · GW(p)

Vanessa Kosoy has a list specifically for her alignment agenda but is probably applicable to agent foundations in general: https://www.alignmentforum.org/posts/fsGEyCYhqs7AWwdCe/learning-theoretic-agenda-reading-list [AF · GW]

No comments

Comments sorted by top scores.