DeepSeek R1 Explained to Grandma

Reinforcement Learning GRPO(D)=E[EQ_θ,ϕE[?] log(π_θ(a|s))] Policy T₀ Aᵢ=Γ·r_mean / s+δev DeepSeek R1 Explained to your grandma My grandma built an AI model in our garage after watching this, thanks bro

A YouTube screenshot shows a man explaining reinforcement learning and a complicated mathematical formula on a whiteboard, captioned as DeepSeek R1 explained to a grandmother. The joke is that the explanation is so advanced and intimidating that the grandmother supposedly builds her own AI model in the garage after watching it. It is used as a sarcastic reaction to overly complicated technical explanations. EN: A YouTube screenshot shows a man explaining reinforcement learning and a complicated mathematical formula on a whiteboard, captioned as DeepSeek R1 explained to a grandmother. The joke is that the explanation is so advanced and intimidating that the grandmother supposedly builds her own AI model in the garage after watching it. It is used as a sarcastic reaction to overly complicated technical explanations.

• @gentoomemes / 15789 2025-01-29