The missing values, appear in most of multivariate time series, prevent advanced analysis of multivariate time series data. Existing imputation approaches try to deal with missing values by deletion, statistical imputation, machine learning based imputation and generative imputation. However, these methods are either incapable of dealing with temporal information or multi-stage. This paper proposes an end-to-end generative model E²GAN to impute missing values in multivariate time series. With the help of the discriminative loss and the squared error loss, E²GAN can impute the incomplete time series by the nearest generated complete time series at one stage. Experiments on multiple real-world datasets show that our model outperforms the baselines on the imputation accuracy and achieves state-of-the-art classification/regression results on the downstream applications. Additionally, our method also gains better time efficiency than multi-stage method on the training of neural networks.
Abstract:Energy harvested from human body movement can produce continuous, stable energy to portable electronics and implanted medical devices. The energy harvesters need to be light, small, inexpensive, and highly portable. Here we report a novel biocompatible device made of poly (vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP) nanofibers on flexible substrates. The nanofibers are prepared with electrospinning followed by a stretching process. This results in aligned nanofibers with diameter control. The assembled device demonstrates high mechanical-to-electrical conversion performance, with stretched PVDF-HFP nanofibers outperforming regular electrospun samples by more than 10 times. Fourier transform infrared spectroscopy (FTIR) reveals that the stretched nanofibers have a higher β phase content, which is the critical polymorph that enables piezoelectricity in polyvinylidene fluoride (PVDF). Polydimethylsiloxane (PDMS) is initially selected as the substrate material for its low cost, high flexibility, and rapid prototyping capability. Bombyx Mori silkworm silk fibroin (SF) and its composites are investigated as promising alternatives due to their high strength, toughness, and biocompatibility. A composite of silk with 20% glycerol demonstrates higher strength and larger ultimate strain than PDMS. With the integration of stretched electrospun PVDF-HFP nanofibers and flexible substrates, this pilot study shows a new pathway for the fabrication of biocompatible, skin-mountable energy devices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.