World Models, JEPA And The Path To Sample-Efficient RL @ycombinator
World Models, JEPA And The Path To Sample-Efficient RL  @ycombinator
Uploaded July 2026 | Updated November 2026, 3 hours ago
In this episode of Decoded, Ankit and Francois walk through the motivation and math behind world models. They cover a number of areas including why sample efficiency is one of the biggest unsolved problems in AI, how deterministic differentiable control and Newtonian physics represent a "perfect world model," and why the action space explosion makes chess tractable but robotics nearly intractable.

Apply to Y Combinator: ycombinator.com/apply
Work at a startup: ycombinator.com/jobs

Transcript: ycrootaccess.substack.com/p/world-models-an-intuitive-introduction

Chapters:

00:00 — Intro
01:45 — What would perfect efficiency look like?
05:10 — World models in the human brain
09:20 — Control theory & the drone example
14:30 — When physics breaks down
17:45 — Chess, Go & the action space problem
24:10 — Why AlphaGo can't scale
28:00 — Monte Carlo tree search explained
34:00 — Self-Driving: state space is infinite
40:30 — Model-Free vs. Model-Based RL
44:00 — Why robotics is the hardest case
48:20 — World models that actually work
54:10 — JEPA & latent space tricks
59:00 — Open problems remaining
01:04:30 — Does this pass the squint test?
01:08:00 — Outro
World Models, JEPA And The Path To Sample-Efficient RLThe New Way To Build A StartupOpen Models Change The Economics of AIAI for GovernmentHow Supabase Became One Of The Fastest Growing DevTool Companies In The WorldAlexandr Wang: “This is a Once-in-a-Civilization Opportunity”How Amplitude Went From Skeptics to “All In” on AI10 Habits PhDs Must Unlearn To Build a Startup | Startup SchoolData for the Real WorldAaron Levie: Why Startups Win In The AI EraAI for Low-Pesticide AgricultureThis Is How You Should Actually Make Launch Videos
Y Combinator |

World Models, JEPA And The Path To Sample-Efficient RL

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER