The International Institute of Forecasters (IIF) has announced that Kin Olivares and Cristian Challu, Ph.D. students in the Machine Learning Department at Carnegie Mellon University, received the 2022 IIF-SAS award for their project proposal in the Methodology category, "Transferability of Neural Forecast Methods."

Founded in 1982, the IIF is dedicated to developing and furthering the generation, distribution and use of forecasting knowledge. The IIF-SAS grant was created to support research on how to improve forecasting methods and business-forecasting practices, including their organizational aspects. The IIF only presents two awards each year to the most promising forecasting research.

Transfer learning involves pretraining flexible models on large datasets and using them for subsequent tasks with minimal additional training. While this technique is widely recognized as a significant achievement in machine learning with numerous practical applications, its use for time series forecasting has not been extensively explored. Olivares and Challu's work to study the transferability of neural forecast methods will allow neural forecasting methods to produce lightning-fast predictions at a fraction of the computational cost. This research will help solve one of the most significant limitations of machine learning methods: the tradeoff between accuracy and speed.

More information about the IIF-SAS grant is available on the IIF website.