auto-encoder

Evolutional Neural Architecture Search for Optimization of Spatiotemporal Brain Network Decomposition

Author(s)3: Qing Li, Wei Zhang, Xia Wu, Tianming Liu
Evolutional Neural Architecture Search for Optimization of Spatiotemporal Brain Network Decomposition 452 480 IEEE Transactions on Biomedical Engineering (TBME)
Using deep neural networks (DNNs) to decompose spatiotemporal brain network has been an important yet challenging problem because the architectures are hard to be designed manually. The previous studies, e.g., deep sparse recurrent auto-encoder (DSRAE), are not optimal in various senses. We employ the evolutionary algorithms to optimize the architecture of DSRAE, named eNAS-DSRAE (i.e., evolutionary Neural Architecture Search on DSRAE). With the validation experiments, our framework can successfully identify the spatiotemporal features and perform better than the hand-crafted DNNs. To our best knowledge, the proposed eNAS-DSRAE is among the earliest NAS models that can extract meaningful spatiotemporal brain networks. read more