Abstract-A multiscale input strategy for multiview deep learning is proposed for supervised multispectral land-use classification and it is validated on a well-known dataset. The hypothesis that simultaneous multiscale views can improve compositionbased inference of classes containing size-varying objects compared to single-scale multiview is investigated. The end-to-end learning system learns a hierarchical feature representation with the aid of convolutional layers to shift the burden of feature determination from hand-engineering to a deep convolutional neural network. This allows the classifier to obtain problemspecific features that are optimal for minimizing the multinomial logistic regression objective, as opposed to user-defined features which trades optimality for generality. A heuristic approach to the optimization of the deep convolutional neural network hyperparameters is used, based on empirical performance evidence. It is shown that a single deep convolutional neural network can be trained simultaneously with multiscale views to improve prediction accuracy over multiple single-scale views. Competitive performance is achieved for the UC Merced dataset where the 93.48% accuracy of multiview deep learning outperforms the 85.37% accuracy of SIFT-based methods and the 90.26% accuracy of unsupervised feature learning.Index Terms-Neural network applications, neural network architecture, feature extraction, urban areas, remote sensing.

Knowledge base question answering (KBQA) is an important task in Natural Language Processing. Existing approaches face significant challenges including complex question understanding, necessity for reasoning, and lack of large end-to-end training datasets. In this work, we propose Neuro-Symbolic Question Answering (NSQA), a modular KBQA system, that leverages (1) Abstract Meaning Representation (AMR) parses for task-independent question understanding; (2) a simple yet effective graph transformation approach to convert AMR parses into candidate logical queries that are aligned to the KB; (3) a pipeline-based approach which integrates multiple, reusable modules that are trained specifically for their individual tasks (semantic parser, entity and relationship linkers, and neuro-symbolic reasoner) and do not require end-to-end training data. NSQA achieves state-of-the-art performance on two prominent KBQA datasets based on DBpedia (QALD-9 and LC-QuAD 1.0). Furthermore, our analysis emphasizes that AMR is a powerful tool for KBQA systems.

Abstract-Multitemporal land-use analysis is becoming increasingly important for the effective management of earth resources. Despite that, consistent differences in the viewingand illumination geometry in satellite-borne imagery introduce some issues in the creation of land-use classification maps. The focus of this study is settlement classification with high-resolution panchromatic acquisitions, using texture features to distinguish between settlement classes. The important multitemporal variance component of shadow is effectively removed before feature determination, which allows for minimum-supervision acrossdate classification. Shadow detection based on local adaptive thresholding is employed and experimentally shown to outperform existing fixed threshold shadow detectors in increasing settlement classification accuracy. Both same and across-date settlement accuracies are significantly improved with shadow masking during feature calculation. A statistical study was performed and found to support the hypothesis that the increased accuracy is due to shadow masking specifically.Index Terms-Image texture analysis, feature extraction, urban areas, remote sensing.

The application of joint source-channel coding, based on fountain codes, in the broadcast timeslot of wireless two-way network coding is investigated in this paper. A computationally efficient subroutine is contributed to the implementation of the fountain compressor, and an error analysis is done. Motivated to develop a true joint source-channel-network code that compresses, adds robustness against channel noise and network codes two packets on a single bipartite graph and iteratively decodes the intended packet on the same Tanner graph, an adaptation of the fountain compressor is presented. The proposed code is shown to outperform a separated joint source-channel and network code in high source entropy and high channel noise regions.

No abstract