City Research Online

Abstract

Multivariate time series have had many applications in areas from healthcare and finance to meteorology and life sciences. Although deep neural networks (DNNs) have shown excellent predictive performance for time series, they have been criticised for being non-interpretable. Neural Additive Models, however, are known to be fully-interpretable by construction, but may achieve far lower predictive performance than DNNs when applied to time series. This work introduces FocusLearn, a fully-interpretable modular neural network capable of matching or surpassing the predictive performance of DNNs trained on multivariate time series. The creation of FocusLearn takes inspiration from modular neural networks and additive models, as well as several experiments aimed at improving the performance of a stand-alone LSTM. In FocusLearn, a recurrent neural network learns the temporal dependencies in the data, while a modified multi-headed attention layer learns to weight selected features while also suppressing redundant features. Modular neural networks are then trained in parallel and independently, one for each selected feature. This modular approach allows the user to inspect how features influence outcomes in the exact same way as with additive models. Experimental results show that this new approach outperforms additive models in both regression and classification of time series tasks, achieving predictive performance that is comparable to state-of-the-art, non-interpretable deep networks applied to time series, and sometimes outperforms them.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Departments:	School of Science & Technology > Department of Computer Science School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads