Signin

How DeepSeek learns: GRPO explained with Triangle Creatures

Dr Mihai Nica • February 23, 2025

Video Thumbnail

You May Also Like

Video Thumbnail

Amateurs Solve a Famous Computer Science Problem On Discord

Quanta Magazine

Video Thumbnail

AI can't cross this line and we don't know why.

Welch Labs

Video Thumbnail

Gradient Descent vs Evolution | How Neural Networks Learn

Emergent Garden

Video Thumbnail

The Art Of Poison-Pilling Music Files

Benn Jordan

Video Thumbnail

I can prove I’ve solved this Sudoku without revealing it

Polylog

Video Thumbnail

What Every Programmer Should Know about How CPUs Work • Matt Godbolt • GOTO 2024

GOTO Conferences

Video Thumbnail

The Boundary of Computation

Mutual Information

Video Thumbnail

Transformers (how LLMs work) explained visually | DL5

3Blue1Brown

Video Thumbnail

Computer Scientist Explains One Concept in 5 Levels of Difficulty | WIRED

WIRED

Video Thumbnail

MIT Introduction to Deep Learning | 6.S191

Alexander Amini

Video Thumbnail

Ai Will Try to Cheat & Escape (aka Rob Miles was Right!) - Computerphile

Computerphile

Video Thumbnail

Training LLM to play chess using Deepseek GRPO reinforcement learning

Efficient NLP

Video Thumbnail

The Biggest Misconception in Physics

Veritasium

Video Thumbnail

Veritasium: What Everyone Gets Wrong About AI and Learning – Derek Muller Explains

Perimeter Institute for Theoretical Physics

Video Thumbnail

How DeepSeek Rewrote the Transformer [MLA]

Welch Labs

Video Thumbnail

[GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Yannic Kilcher

Video Thumbnail

The Man Who Almost Broke Math (And Himself...)

Veritasium

Video Thumbnail

Stand-up Maths' max-of-dice conjecture proven! Calculating the average maximum of dice rolls

Dr Mihai Nica

Video Thumbnail

Everything We Know About 'Oumuamua

Astrum

Video Thumbnail

Visualizing transformers and attention | Talk for TNG Big Tech Day '24

Grant Sanderson

Video Thumbnail

Jensen Huang on GPUs - Computerphile

Computerphile

Video Thumbnail

DeepSeek R1 Theory Overview | GRPO + RL + SFT

Deep Learning with Yacine

Video Thumbnail

THIS Is Why Gen Z Are The Poorest Generation

The Infographics Show

Video Thumbnail

Generative Model That Won 2024 Nobel Prize

Artem Kirsanov

Video Thumbnail

Is Human Data Enough? With David Silver

Google DeepMind

Dr Mihai Nica

About

No channel description available.

Latest Posts

Video Thumbnail

"A Random Variable is NOT Random and NOT a Variable"

Dr Mihai Nica

Video Thumbnail

Stand-up Maths' max-of-dice conjecture proven! Calculating the average maximum of dice rolls

Dr Mihai Nica

Video Thumbnail

How DeepSeek learns: GRPO explained with Triangle Creatures

Dr Mihai Nica

Video Thumbnail

The Coin Flip Game that Stumped Twitter: Alice HH vs Bob HT

Dr Mihai Nica