Multi-Task Dynamical Systems
This addresses the need for improved time series modeling in domains like healthcare and motion analysis, though it appears incremental as an extension of multi-task learning to time series.
The paper tackles the problem of modeling time series data from multiple entities by introducing the multi-task dynamical system (MTDS), a method that allows models to specialize to individual sequences while sharing commonalities across them, applied to motion-capture and patient drug-response data.
Time series datasets are often composed of a variety of sequences from the same domain, but from different entities, such as individuals, products, or organizations. We are interested in how time series models can be specialized to individual sequences (capturing the specific characteristics) while still retaining statistical power by sharing commonalities across the sequences. This paper describes the multi-task dynamical system (MTDS); a general methodology for extending multi-task learning (MTL) to time series models. Our approach endows dynamical systems with a set of hierarchical latent variables which can modulate all model parameters. To our knowledge, this is a novel development of MTL, and applies to time series both with and without control inputs. We apply the MTDS to motion-capture data of people walking in various styles using a multi-task recurrent neural network (RNN), and to patient drug-response data using a multi-task pharmacodynamic model.