In this paper, the problem of the cross-domain fault diagnosis of rotating machinery is considered. In a practical setting of this approach, the operating platform of the machine may have a different setup and conditions compared to the experimental platform that is used to collect the training data. This can lead to significant data variations, specifically, domain shifts. Conventional data-driven approaches are known to adapt poorly to these domain shifts, resulting in a significant drop in the diagnosis accuracy when the pre-trained model is applied in the actual operating situation. In this paper, an unsupervised domain adaptation approach is developed to mitigate the domain shifts between the data gathered from the experimental platform (the source domain) and the operating platform (the target domain) by aligning the features extracted from the two data domains. The mutual information between the target feature space and the entire feature space is maximized to improve the knowledge transferability of the labeled data in the source domain. Furthermore, the feature-level discrepancy between the two domains is minimized to further improve diagnosis accuracy. The experiments using public datasets and real-world adaptation scenarios demonstrate the feasibility and the superior performance of the proposed method.

Accurately predicting human locomotion intent is beneficial in controlling wearable robots and in assisting humans to walk smoothly on different terrains. Traditional methods for predicting human locomotion intent require collecting and labeling the human signals, and training specific classifiers for each new subject, which introduce a heavy burden on both the subject and the researcher. In addressing this issue, the present study liberates the subject and the researcher from labeling a large amount of data, by incorporating an unsupervised cross-subject adaptation method to predict the locomotion intent of a target subject whose signals are not labeled. The adaptation is realized by designing two classifiers to maximize the classification discrepancy and a feature generator to align the hidden features of the source and the target subjects to minimize the classification discrepancy. A neural network is trained by the labeled training set of source subjects and the unlabeled training set of target subjects. Then it is validated and tested on the validation set and the test set of target subjects. Experimental results in the leave-one-subject-out test indicate that the present method can classify the locomotion intent and activities of target subjects at the averaged accuracy of 93.60% and 94.59% on two public datasets. The present method increases the user-independence of the classifiers, but it has been evaluated only on the data of subjects without disabilities. The potential of the present method to predict the locomotion intent of subjects with disabilities and control the wearable robots will be evaluated in future work.

Developing compute platforms capable of performing computations at high speed is essential for data processing in the next generation of data centers and edge devices. A neuromorphic photonic accelerator on a silicon photonic platform is a promising solution. Compared to silicon photonic data communication transceiver modules, neuromorphic photonic accelerators constitute a large number of active and passive components and optoelectronic devices to handle the parallel processing. Thus, an increased number of optical and electrical interconnects are required, making the packaging of such processors challenging. Moreover, thermal and electrical crosstalk can dramatically degrade the performance of such processors. Thus, packaging a neuromorphic photonic accelerator for efficient processing and data movement requires careful considerations at the chip, module, and board levels. This work investigates the challenges and potential solutions for optical coupling, optical and electrical interconnections, processor-memory communication, and thermal and electrical cross-talk to develop neuromorphic photonic accelerators.

This paper addresses the problem of optimizing sensor deployment locations of a wireless sensor network to reconstruct and also predict a spatiotemporal signal. Traditional sensor deployment optimization approaches only selects deployment locations and provide the feature in the sampled area. Hence, if the sampling resources are limited, the deployed sensors may not be able to provide all features in the field of interest (input space). To solve this issue, a deep learning framework is developed to reconstruct and predict the entire spatiotemporal signal from a limited amount of observations. In the beginning, the proposed approach optimizes sampling locations to retrieve sufficient features for reconstruction and prediction from historical data. Then, a spatiotemporal autoencoder is used to capture the nonlinear mappings from the in-situ measurements to the entire spatiotemporal signal. A simulation is conducted using global climate datasets from the National Oceanic and Atmospheric Administration, to implement and validate the developed methodology. The results demonstrate a significant improvement made by the proposed algorithm. Specifically, compared to traditional approaches, the proposed method provides superior performance in terms of both reconstruction error and spatial prediction robustness.

In this article, the unsupervised domain adaptation problem, where an approximate inference model is to be learned from a labeled dataset and expected to generalize well on an unlabeled dataset, is considered. Unlike the existing work, we explicitly unveil the importance of the latent variables produced by the feature extractor, that is, encoder, where contains the most representative information about their input samples, for the knowledge transfer. We argue that an estimator of the representation of the two datasets can be used as an agent for knowledge transfer. To be specific, a novel variational inference approach is proposed to approximate a latent distribution from the unlabeled dataset that can be used to accurately predict its input samples. It is demonstrated that the discriminative knowledge of the latent distribution that is learned from the labeled dataset can be progressively transferred to that is learned from the unlabeled dataset by simultaneously optimizing the estimator via the variational inference and our proposed regularization for shifting the mean of the estimator. The experiments on several benchmark datasets demonstrate that the proposed method consistently outperforms state-of-the-art methods for both object classification and digit classification.