← Volver a NERDflix

Forecasting & Large Language Models (LLMs)

Protagonista: Matias Callara
Año: 2024
País: Argentina
Género: Misterio
Track: Data Science / Ai
Idioma: Español

In this session, we explore the latest developments in using Large Language Models (LLMs) for time series forecasting. We cover techniques ranging from general-use LLMs like GPT-4 and Llama 3, using methods like PromptCast or LLMTime, to specialized foundation models designed specifically for time series forecasting, like Amazon's Chronos and Google's TimesFM. We discuss the performance of these models through benchmarks conducted by members of the forecasting community on open datasets. This presentation is designed for a general audience without specific forecasting experience, providing an update on the ongoing work in this area and highlighting contributions from various teams. Join us to gain insights into current trends and advancements in the field of time series forecasting.

data-science-ai ai llms forecasting

Forecasting & Large Language Models (LLMs)

Sobre Matias Callara

Matias Callara

Senior Data Scientist

Matias Callara is an Industrial Engineer with a Master’s in Optimization and a Ph.D. in Informatics. With over 10 years of experience at the intersection of Machine Learning, Operational Research, and Decision Theory, Matias has designed and led the development of Roche Diagnostics’ Advanced Analytics Forecasting System. He is currently a Senior Data Scientist at Roche, focusing on high-value strategic AI initiatives, and is also an invited professor at Torcuato Di Tella University.

𝕏 Twitter LinkedIn