Efficient learning of nonlinear prediction models with time-series privileged information
Paper i proceeding, 2022

In domains where sample sizes are limited, efficient learning algorithms are critical. Learning using privileged information (LuPI) offers increased sample efficiency by allowing prediction models access to types of information at training time which is unavailable when the models are used. In recent work, it was shown that for prediction in linear-Gaussian dynamical systems, a LuPI learner with access to intermediate time series data is never worse and often better in expectation than any unbiased classical learner. We provide new insights into this analysis and generalize it to nonlinear prediction tasks in latent dynamical systems, extending theoretical guarantees to the case where the map connecting latent variables and observations is known up to a linear transform. In addition, we propose algorithms based on random features and representation learning for the case when this map is unknown. A suite of empirical results confirm theoretical findings and show the potential of using privileged time-series information in nonlinear prediction.


Bastian Jung

Student vid Chalmers

Fredrik Johansson

Chalmers, Data- och informationsteknik, Data Science och AI

Advances in Neural Information Processing Systems 35 (NeurIPS 2022)

Thirty-seventh Conference on Neural Information Processing Systems
New Orleans, USA,

WASP AI/MLX Forskarassistent

Wallenberg AI, Autonomous Systems and Software Program, 2019-08-01 -- 2023-08-01.


Informations- och kommunikationsteknik


Sannolikhetsteori och statistik

Datavetenskap (datalogi)

Mer information

Senast uppdaterat