Naval Uses Bike-Riding Analogy to Explain SFT vs RL

Naval shared an ELI5 explanation of DeepSeek's training, comparing SFT to giving a child a bike manual and RL to letting them learn by falling and retrying, arguing RL helps intelligence truly grow.

2026-09-06 ~ 2026-09-07 · 2 related posts

• Naval Amplifies DeepSeek Explainer: SFT Is a Bike Manual, RL Is Learning to Ride — McDonaghMatthew · 2026-09-06
• Naval's bike analogy for SFT vs RL explains why DeepSeek R1 shocked the industry — McDonaghMatthew · 2026-09-07