National University of Defense Technology

About: National University of Defense Technology is a education organization based out in Changsha, China. It is known for research contribution in the topics: Computer science & Radar. The organization has 39430 authors who have published 40181 publications receiving 358979 citations. The organization is also known as: Guófáng Kēxuéjìshù Dàxué & NUDT.

GRAINS: Generative Recursive Autoencoders for INdoor Scenes

Abstract: We present a generative neural network that enables us to generate plausible 3D indoor scenes in large quantities and varieties, easily and highly efficiently. Our key observation is that indoor scene structures are inherently hierarchical. Hence, our network is not convolutional; it is a recursive neural network, or RvNN. Using a dataset of annotated scene hierarchies, we train a variational recursive autoencoder, or RvNN-VAE, which performs scene object grouping during its encoding phase and scene generation during decoding. Specifically, a set of encoders are recursively applied to group 3D objects based on support, surround, and co-occurrence relations in a scene, encoding information about objects’ spatial properties, semantics, and relative positioning with respect to other objects in the hierarchy. By training a variational autoencoder (VAE), the resulting fixed-length codes roughly follow a Gaussian distribution. A novel 3D scene can be generated hierarchically by the decoder from a randomly sampled code from the learned distribution. We coin our method GRAINS, for Generative Recursive Autoencoders for INdoor Scenes. We demonstrate the capability of GRAINS to generate plausible and diverse 3D indoor scenes and compare with existing methods for 3D scene synthesis. We show applications of GRAINS including 3D scene modeling from 2D layouts, scene editing, and semantic scene segmentation via PointNet whose performance is boosted by the large quantity and variety of 3D scenes generated by our method.

Reinforced Mnemonic Reader for Machine Reading Comprehension

Abstract: In this paper, we introduce the Reinforced Mnemonic Reader for machine reading comprehension tasks, which enhances previous attentive readers in two aspects. First, a reattention mechanism is proposed to refine current attentions by directly accessing to past attentions that are temporally memorized in a multi-round alignment architecture, so as to avoid the problems of attention redundancy and attention deficiency. Second, a new optimization approach, called dynamic-critical reinforcement learning, is introduced to extend the standard supervised method. It always encourages to predict a more acceptable answer so as to address the convergence suppression problem occurred in traditional reinforcement learning algorithms. Extensive experiments on the Stanford Question Answering Dataset (SQuAD) show that our model achieves state-of-the-art results. Meanwhile, our model outperforms previous systems by over 6% in terms of both Exact Match and F1 metrics on two adversarial SQuAD datasets.

Data Processing and Text Mining Technologies on Electronic Medical Records: A Review.

Abstract: Currently, medical institutes generally use EMR to record patient’s condition, including diagnostic information, procedures performed, and treatment results. EMR has been recognized as a valuable resource for large-scale analysis. However, EMR has the characteristics of diversity, incompleteness, redundancy, and privacy, which make it difficult to carry out data mining and analysis directly. Therefore, it is necessary to preprocess the source data in order to improve data quality and improve the data mining results. Different types of data require different processing technologies. Most structured data commonly needs classic preprocessing technologies, including data cleansing, data integration, data transformation, and data reduction. For semistructured or unstructured data, such as medical text, containing more health information, it requires more complex and challenging processing methods. The task of information extraction for medical texts mainly includes NER (named-entity recognition) and RE (relation extraction). This paper focuses on the process of EMR processing and emphatically analyzes the key techniques. In addition, we make an in-depth study on the applications developed based on text mining together with the open challenges and research issues for future work.

Deep Convolutional Neural Network-Based Early Automated Detection of Diabetic Retinopathy Using Fundus Image.

Abstract: The automatic detection of diabetic retinopathy is of vital importance, as it is the main cause of irreversible vision loss in the working-age population in the developed world. The early detection of diabetic retinopathy occurrence can be very helpful for clinical treatment; although several different feature extraction approaches have been proposed, the classification task for retinal images is still tedious even for those trained clinicians. Recently, deep convolutional neural networks have manifested superior performance in image classification compared to previous handcrafted feature-based image classification methods. Thus, in this paper, we explored the use of deep convolutional neural network methodology for the automatic classification of diabetic retinopathy using color fundus image, and obtained an accuracy of 94.5% on our dataset, outperforming the results obtained by using classical approaches.

Spatial Joins Using R-trees: Breadth-First Traversal with Global Optimizations

Abstract: R-tree based spatial join is useful because of both its superior performance and the wide spread implementation of R-trees. We present a new R-tree join method called BFRJ (Breadth-First R-tree Join). BFRJ synchronously traverses both R-trees in breadthfirst order while processing join computation one level at a time. At each level, BFRJ creates an intermediate join index and deploys global optimization strategies (ordering, memory management, buffer management) to improve the join computation at the next level. We also present an experimental evaluation of the proposed optimizations as well as a performance comparison between BFRJ and the state-of-the-art approach. Our experimental results indicate that BFRJ with global optimizations can outperform the competitor by a significant margin (up to 50%). This work was supportedin part by the University of Michigan ITS Research Center of Excellence grant (DTFH61-93-X0001i’-Sub) sponsored by the U.S. Dept. of Transportation and by the Michigan Dept. of Transportation. This work was performed while the authors were at the University of Michigan. Permission to copy without fee all OT part of this material is granted provided that the copies are not made OT distributed for direct commercial advantage, the VLDB copyright notice and the title of the publication and its date appear, and notice is given that copying is by permission of the Very Large Data Base Endowment. To copy otherwise, OT to republish, requires a fee and/or special permission from the Endowment. Proceedings of the 23rd VLDB Conference Athens, Greece, 1997