Vai al contenuto
AI.info

events

Marketing Foundation Models Bayesian MMM Without Per-Dataset Fitting

I ricercatori di PyMC Labs presentano Marketing Foundation Models, un prototipo che stima i contributi del marketing mix in un solo passaggio, senza adattamento ai singoli dataset.

Marketing Foundation Models Bayesian MMM Without Per-Dataset Fitting | Eventi sull’IA

In inglese

Starts
2026-11-17 15:00 UTC
Organizer
PyMC Labs Online Meetup
Location
Online event

PyMC Labs hosts an online session on Marketing Foundation Models, a research prototype that estimates baseline, channel and control contributions in one pass, without fitting a model to each new dataset. Carlos Trujillo and Teemu Säilynoja describe how it combines a simulator of marketing worlds, a transformer encoder and flow matching, and what synthetic evaluation showed, including uncertainty calibration. Luca Fiaschi hosts, and the session covers the prototype's limits and how it differs from TabPFN and TimesFM. It targets marketing analytics teams, Bayesian modelers and analytics leaders.

Event website