ConFuse: Convolutional Transform Learning Fusion Framework For Multi-Channel Data Analysis

11/09/2020

∙

by   Pooja Gupta, et al.

∙

0

∙

share

This work addresses the problem of analyzing multi-channel time series data processed by a separate 1D convolutional transform; the output of all the channels are fused by a fully connected layer of transform learning. The training procedure takes advantage of the proximal interpretation of activation functions. We apply the developed framework to multi-channel financial data for stock forecasting and trading. We compare our proposed formulation with benchmark deep time series analysis networks. The results show that our method yields considerably better results than those compared against.