Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Deep Factors for Forecasting

About

Producing probabilistic forecasts for large collections of similar and/or dependent time series is a practically relevant and challenging task. Classical time series models fail to capture complex patterns in the data, and multivariate techniques struggle to scale to large problem sizes. Their reliance on strong structural assumptions makes them data-efficient, and allows them to provide uncertainty estimates. The converse is true for models based on deep neural networks, which can learn complex patterns and dependencies given enough data. In this paper, we propose a hybrid model that incorporates the benefits of both approaches. Our new method is data-driven and scalable via a latent, global, deep component. It also handles uncertainty through a local classical model. We provide both theoretical and empirical evidence for the soundness of our approach through a necessary and sufficient decomposition of exchangeable time series into a global and a local part. Our experiments demonstrate the advantages of our model both in term of data efficiency, accuracy and computational complexity.

Yuyang Wang, Alex Smola, Danielle C. Maddix, Jan Gasthaus, Dean Foster, Tim Januschowski• 2019

Related benchmarks

TaskDatasetResultRank
Probabilistic time series forecastingBitcoin, COVID-19, and Synthetic Average (test)
CRPS0.329
12
Showing 1 of 1 rows

Other info

Follow for update