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.

Open-Domain Targeted Sentiment Analysis via Span-Based Extraction and Classification

Abstract: Open-domain targeted sentiment analysis aims to detect opinion targets along with their sentiment polarities from a sentence. Prior work typically formulates this task as a sequence tagging problem. However, such formulation suffers from problems such as huge search space and sentiment inconsistency. To address these problems, we propose a span-based extract-then-classify framework, where multiple opinion targets are directly extracted from the sentence under the supervision of target span boundaries, and corresponding polarities are then classified using their span representations. We further investigate three approaches under this framework, namely the pipeline, joint, and collapsed models. Experiments on three benchmark datasets show that our approach consistently outperforms the sequence tagging baseline. Moreover, we find that the pipeline model achieves the best performance compared with the other two models.

Joint 2D-DOD, 2D-DOA, and Polarization Angles Estimation for Bistatic EMVS-MIMO Radar via PARAFAC Analysis

Abstract: Multiple-input multiple-output (MIMO) is a technical hotspot in physical layer with numerous applications in wireless communications, radars, sonars, and well beyond. In this paper, we focus on the multi-dimensional angle estimation problem in a bistatic electromagnetic vector sensors (EMVS) MIMO system. Namely, we need to simultaneously estimate two-dimensional (2D) direction-of-arrival (DOA), 2D direction-of-departure (DOD), 2D receive polarization angle (RPA) and 2D transmit polarization angle (TPA). To tackle this issue, a parallel factor (PARAFAC) analysis-based estimator is proposed. Firstly, a third-order PARAFAC analysis data model is established, which can efficiently exploit the tensor structure of the array measurement. After performing PARAFAC decomposition on the tensor measurement, the factor matrices are achieved. By combining the estimation method of signal parameters via rotational invariance technique (ESPRIT) with the vector cross-product method, joint estimates of 2D-DOD, 2D-DOA, 2D-TPA and 2D-RPA are obtained without further pairing calculation. Compared with the state-of-the-art ESPRIT-Like approach, the proposed method can achieve better performance by enforcing the third-order structure information, and it is suitable for arbitrary array manifolds. Theoretical analyses are given and numerical results corroborate our analysis.

Whole packet forwarding: Efficient design of fully adaptive routing algorithms for networks-on-chip

Abstract: Routing algorithms for networks-on-chip (NoCs) typically only have a small number of virtual channels (VCs) at their disposal. Limited VCs pose several challenges to the design of fully adaptive routing algorithms. First, fully adaptive routing algorithms based on previous deadlock-avoidance theories require a conservative VC re-allocation scheme: a VC can only be re-allocated when it is empty, which limits performance. We propose a novel VC re-allocation scheme, whole packet forwarding (WPF), which allows a non-empty VC to be re-allocated. WPF leverages the observation that the majority of packets in NoCs are short. We prove that WPF does not induce deadlock if the routing algorithm is deadlock-free using conservative VC re-allocation. WPF is an important extension of previous deadlock-avoidance theories. Second, to efficiently utilize WPF in VC-limited networks, we design a novel fully adaptive routing algorithm which maintains packet adaptivity without significant hardware cost. Compared with conservative VC re-allocation, WPF achieves an average 88.9% saturation throughput improvement in synthetic traffic patterns and an average 21.3% and maximal 37.8% speedup for PARSEC applications with heavy network loads. Our design also offers higher performance than several partially adaptive and deterministic routing algorithms.1

A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoning

Abstract: Rapid progress has been made in the field of reading comprehension and question answering, where several systems have achieved human parity in some simplified settings. However, the performance of these models degrades significantly when they are applied to more realistic scenarios, such as answers involve various types, multiple text strings are correct answers, or discrete reasoning abilities are required. In this paper, we introduce the Multi-Type Multi-Span Network (MTMSN), a neural reading comprehension model that combines a multi-type answer predictor designed to support various answer types (e.g., span, count, negation, and arithmetic expression) with a multi-span extraction method for dynamically producing one or multiple text spans. In addition, an arithmetic expression reranking mechanism is proposed to rank expression candidates for further confirming the prediction. Experiments show that our model achieves 79.9 F1 on the DROP hidden test set, creating new state-of-the-art results. Source code (https://github.com/huminghao16/MTMSN) is released to facilitate future work.

Multiple-Spectral-Band CRFs for Denoising Junk Bands of Hyperspectral Imagery

Abstract: Denoising of hyperspectral imagery in the domain of imaging spectroscopy by conditional random fields (CRFs) is addressed in this work. For denoising of hyperspectral imagery, the strong dependencies across spatial and spectral neighbors have been proved to be very useful. Many available hyperspectral image denoising algorithms adopt multidimensional tools to deal with the problems and thus naturally focus on the use of the spectral dependencies. However, few of them were specifically designed to use the spatial dependencies. In this paper, we propose a multiple-spectral-band CRF (MSB-CRF) to simultaneously model and use the spatial and spectral dependencies in a unified probabilistic framework. Furthermore, under the proposed MSB-CRF framework, we develop two hyperspectral image denoising algorithms, which, thanks to the incorporated spatial and spectral dependencies, can significantly remove the noise, while maintaining the important image details. The experiments are conducted in both simulated and real noisy conditions to test the proposed denoising algorithms, which are shown to outperform the popular denoising methods described in the previous literatures.