Lag-Llama: Towards Foundation Models for Probabilistic Time Series Forecasting

Welcome to the AI research bites. This series of short and informative talks showcases cutting-edge research work from ServiceNow AI Research team. The AI Research Bites are open to all, especially those interested in keeping up with the fast-paced AI research community.
In this session, Arjun Ashok presents Lag-Llama, the first open-source foundation model for probabilistic time series forecasting. Lag-Llama showcases strong zero-shot performance and is state-of-the-art in probabilistic univariate forecasting on finetuning.
Paper: arxiv.org/abs/2310.08278
Code, Models and Demo: github.com/time-series-founda...
ServiceNow AI Research team: www.servicenow.com/research/

[@AgostinoLurani]

Is it possible to fine-tune using covariate ts as well?

[@pauljones9150]

Whats the time complexity of this model? How does it scale?