#91 - HATTIE ZHOU - Teaching Algorithmic Reasoning via In-context Learning #NeurIPS

#91 - HATTIE ZHOU - Teaching Algorithmic Reasoning via In-context Learning #NeurIPS

Support us! https://www.patreon.com/mlst


Hattie Zhou, a PhD student at Université de Montréal and Mila, has set out to understand and explain the performance of modern neural networks, believing it a key factor in building better, more trusted models. Having previously worked as a data scientist at Uber, a private equity analyst at Radar Capital, and an economic consultant at Cornerstone Research, she has recently released a paper in collaboration with the Google Brain team, titled ‘Teaching Algorithmic Reasoning via In-context Learning’. In this work, Hattie identifies and examines four key stages for successfully teaching algorithmic reasoning to large language models (LLMs): formulating algorithms as skills, teaching multiple skills simultaneously, teaching how to combine skills, and teaching how to use skills as tools. Through the application of algorithmic prompting, Hattie has achieved remarkable results, with an order of magnitude error reduction on some tasks compared to the best available baselines. This breakthrough demonstrates algorithmic prompting’s viability as an approach for teaching algorithmic reasoning to LLMs, and may have implications for other tasks requiring similar reasoning capabilities.


TOC

[00:00:00] Hattie Zhou

[00:19:49] Markus Rabe [Google Brain]


Hattie's Twitter - https://twitter.com/oh_that_hat

Website - http://hattiezhou.com/


Teaching Algorithmic Reasoning via In-context Learning [Hattie Zhou, Azade Nova, Hugo Larochelle, Aaron Courville, Behnam Neyshabur, and Hanie Sedghi]

https://arxiv.org/pdf/2211.09066.pdf


Markus Rabe [Google Brain]:

https://twitter.com/markusnrabe

https://research.google/people/106335/

https://www.linkedin.com/in/markusnrabe


Autoformalization with Large Language Models [Albert Jiang Charles Edgar Staats Christian Szegedy Markus Rabe Mateja Jamnik Wenda Li Yuhuai Tony Wu]

https://research.google/pubs/pub51691/


Discord: https://discord.gg/aNPkGUQtc5

YT: https://youtu.be/80i6D2TJdQ4

Denne episoden er hentet fra en åpen RSS-feed og er ikke publisert av Podme. Den kan derfor inneholde annonser.

Episoder(260)

Stealing Reasoning Traces from Proprietary LLM APIs — Ilia Shumailov & Alexander Panfilov

Stealing Reasoning Traces from Proprietary LLM APIs — Ilia Shumailov & Alexander Panfilov

Tim Scarfe speaks with Ilia Shumailov and Alexander Panfilov about their paper, Stealing Reasoning Traces from Proprietary LLM APIs.The core bug sounds deceptively simple: providers return encrypted r...

22 Aug 49min

Every Exponential Ends — Silicon Valley Forgot — Adam Becker

Every Exponential Ends — Silicon Valley Forgot — Adam Becker

Astrophysicist Adam Becker, author of "What Is Real?", joins Tim Scarfe to take apart the futures Silicon Valley keeps selling: the 2045 singularity, mind uploading, Mars colonies, and the AI apocalyp...

20 Aug 1h 18min

AI Is Learning at the Wrong Level of Abstraction — Matthieu Wyart

AI Is Learning at the Wrong Level of Abstraction — Matthieu Wyart

This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlstWhy can deep networks discover abstractions that shallow models miss? Statistical phys...

10 Aug 1h 18min

How Researchers Test AI for Hidden Goals — Apollo Research

How Researchers Test AI for Hidden Goals — Apollo Research

Can an AI do the right thing for the wrong reason? Tim Scarfe speaks with Apollo Research’s Alexander Meinke, Axel Højmark and Jérémy Scheurer about Measuring Reward-Seeking via Contrastive Belief Upd...

31 Jul 1h 18min

Why a Nation Can't Outsource Its Frontier AI - Alistair Pullen (Cosine AI)

Why a Nation Can't Outsource Its Frontier AI - Alistair Pullen (Cosine AI)

This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlstBritain's most capable coding model can't be exported, and that ban is the whole reaso...

13 Jul 55min

 The Benchmark With No Instructions — ARC-AGI-3 (winning team!)

The Benchmark With No Instructions — ARC-AGI-3 (winning team!)

Tim Scarfe travels to Zurich to sit down with the Tufa Labs ARC-AGI-3 team — founder Benjamin Crouzier, with Jeroen Cottaar, Dries Smit, Stefano Viel and Michal Tesnar — to work out what their leaderb...

1 Jul 1h 24min

The Thermodynamic AI Computing Chip - Thomas Ahle

The Thermodynamic AI Computing Chip - Thomas Ahle

Thomas Ahle wants Normal Computing to be the Lovable for chip design: type your intent, and a swarm of agents carries it from design through optimisation, formalisation and verification to tape-out. T...

28 Jun 1h 2min

He won a Nobel here for AlphaFold. Then he left. - John Jumper

He won a Nobel here for AlphaFold. Then he left. - John Jumper

This episode is sponsored by Notion. Learn more about Notion's Developer Platform today at https://notion.com/mlstProtein folding stalled biology for fifty years. A sequence of amino acids dictates a ...

22 Jun 53min

Populært innen Teknologi

lydartikler-fra-aftenposten
teknisk-sett
tomprat-med-gunnar-tjomlid
smart-forklart
rss-kunstig-intelligens-med-elisabeth-maren-og-morten
nasjonal-sikkerhetsmyndighet-nsm
shifter
elektropodden
rss-alt-som-gar-pa-strom
energi-og-klima
fornybaren
rss-snakk-om-sikkerhet
rss-ki-praten
teknologi-og-mennesker
rss-ai-forklart
pedagogisk-intelligens
rss-heis
rss-fisketimen
rss-bak-skyen
rss-god-blanding