LLM hallucinations discover new math solutions!? | FunSearch explained

AI Coffee Break with Letitia
AI Coffee Break with Letitia
12.4 هزار بار بازدید - 8 ماه پیش - The unreasonable effectiveness of guided
The unreasonable effectiveness of guided confabulation: Solving math problems with hallucinatory LLMs is now possible!? 🤯 We explain how Google DeepMind did it.
Bonus: The answer of Fields Medalist 2022 Hugo Duminil-Copin at the last #HLF23 after I asked about AI helping mathematicians solve problems.
➡️ AI Coffee Break Merch! 🛍️ https://aicoffeebreak.creator-spring....

📚 FunSearch blog post: https://deepmind.google/discover/blog...

📜 “Mathematical discoveries from program search with large language models” by Bernardino Romera-Paredes, Mohammadamin Barekatain, Alexander Novikov, Matej Balog, M. Pawan Kumar, Emilien Dupont, Francisco J. R. Ruiz, Jordan S. Ellenberg, Pengming Wang, Omar Fawzi, Pushmeet Kohli & Alhussein Fawzi, Nature 2023 https://www.nature.com/articles/s4158...

🕊️ “Mathematics will fall first” Twitter: 1731096582932578653

Thanks to our Patrons who support us in Tier 2, 3, 4: 🙏
Dres. Trost GbR, Siltax, Vignesh Valliappan, ‪@Mutual_Information‬ ,  Kshitij

Outline:
00:00 Mathematics will fall first
01:19 time computation / Guided confabulation
03:02 FunSearch explained
05:14 Cap Set problem
06:23 Bin Packing problem
07:22 Mathematicians say this
10:42 Will AI take mathematician’s jobs?

▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀
🔥 Optionally, pay us a coffee to help with our Coffee Bean production!  ☕
Patreon: Patreon: AICoffeeBreak
Ko-fi: https://ko-fi.com/aicoffeebreak
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀

🔗 Links:
AICoffeeBreakQuiz: aicoffeebreak
Twitter: Twitter: AICoffeeBreak
Reddit: Reddit: AICoffeeBreak
YouTube: aicoffeebreak

#AICoffeeBreak #MsCoffeeBean #MachineLearning #AI #research​
Mushroom in the thumbnail generated with DALL-E (GPT-4 interface)

Video editing: Nils Trost
8 ماه پیش در تاریخ 1402/09/27 منتشر شده است.
12,481 بـار بازدید شده
... بیشتر