Zhishuo ZHANG Chengxiang TAN Xueyan ZHAO Min YANG
Entity alignment (EA) is a crucial task for integrating cross-lingual and cross-domain knowledge graphs (KGs), which aims to discover entities referring to the same real-world object from different KGs. Most existing embedding-based methods generate aligning entity representation by mining the relevance of triple elements, paying little attention to triple indivisibility and entity role diversity. In this paper, a novel framework named TTEA - Type-enhanced Ensemble Triple Representation via Triple-aware Attention for Cross-lingual Entity Alignment is proposed to overcome the above shortcomings from the perspective of ensemble triple representation considering triple specificity and diversity features of entity role. Specifically, the ensemble triple representation is derived by regarding relation as information carrier between semantic and type spaces, and hence the noise influence during spatial transformation and information propagation can be smoothly controlled via specificity-aware triple attention. Moreover, the role diversity of triple elements is modeled via triple-aware entity enhancement in TTEA for EA-oriented entity representation. Extensive experiments on three real-world cross-lingual datasets demonstrate that our framework makes comparative results.
Ryuta TAMURA Yuichi TAKANO Ryuhei MIYASHIRO
We study the mixed-integer optimization (MIO) approach to feature subset selection in nonlinear kernel support vector machines (SVMs) for binary classification. To measure the performance of subset selection, we use the distance between two classes (DBTC) in a high-dimensional feature space based on the Gaussian kernel function. However, DBTC to be maximized as an objective function is nonlinear, nonconvex and nonconcave. Despite the difficulty of linearizing such a nonlinear function in general, our major contribution is to propose a mixed-integer linear optimization (MILO) formulation to maximize DBTC for feature subset selection, and this MILO problem can be solved to optimality using optimization software. We also derive a reduced version of the MILO problem to accelerate our MILO computations. Experimental results show good computational efficiency for our MILO formulation with the reduced problem. Moreover, our method can often outperform the linear-SVM-based MILO formulation and recursive feature elimination in prediction performance, especially when there are relatively few data instances.
Wang XU Yongliang MA Kehai CHEN Ming ZHOU Muyun YANG Tiejun ZHAO
Non-autoregressive generation has attracted more and more attention due to its fast decoding speed. Latent alignment objectives, such as CTC, are designed to capture the monotonic alignments between the predicted and output tokens, which have been used for machine translation and sentence summarization. However, our preliminary experiments revealed that CTC performs poorly on document abstractive summarization, where a high compression ratio between the input and output is involved. To address this issue, we conduct a theoretical analysis and propose Hierarchical Latent Alignment (HLA). The basic idea is a two-step alignment process: we first align the sentences in the input and output, and subsequently derive token-level alignment using CTC based on aligned sentences. We evaluate the effectiveness of our proposed approach on two widely used datasets XSUM and CNNDM. The results indicate that our proposed method exhibits remarkable scalability even when dealing with high compression ratios.
Conggai LI Feng LIU Xin ZHOU Yanli XU
To obtain a full picture of potential applications for propagation-delay based X channels, it is important to obtain all feasible schemes of cyclic interference alignment including the encoder, channel instance, and decoder. However, when the dimension goes larger, theoretical analysis about this issue will become tedious and even impossible. In this letter, we propose a computer-aided solution by searching the channel space and the scheduling space, which can find all feasible schemes in details. Examples are given for some typical X channels. Computational complexity is further analyzed.
Conggai LI Qian GAN Feng LIU Yanli XU
Compared with the unicast scenario, X channels with multicast messaging can support richer transmission scenarios. The transmission efficiency of the wireless multicast X channel is an important and open problem. This article studies the degrees of freedom of a propagation-delay based multicast X channel with two transmitters and arbitrary receivers, where each transmitter sends K different messages and each receiver desires K - 1 of them from each transmitter. The cyclic polynomial approach is adopted for modeling and analysis. The DoF upper bound is analyzed and shown to be unreachable. Then a suboptimal scheme with one extra time-slot cycle is proposed, which uses the cyclic interference alignment method and achieves a DoF of K - 1. Finally, the feasibility conditions in the Euclidean space are derived and the potential applications are demonstrated for underwater acoustic and terrestrial radio communications.
Masatoshi YAITA Yosei SHIBATA Takahiro ISHINABE Hideo FUJIKAKE
In this paper, we proposed the phase disturbing device using randomly-fluctuated liquid crystal (LC) alignment to reduce the speckle noise generated in holographic displays. Some parameters corresponding to the alignment fluctuation of thick LC layer were quantitatively evaluated, and we clarified the effect of the LC alignment fluctuation with the parameters on speckle noise reduction.
Daming LIN Jie WANG Yundong LI
Rapid building damage identification plays a vital role in rescue operations when disasters strike, especially when rescue resources are limited. In the past years, supervised machine learning has made considerable progress in building damage identification. However, the usage of supervised machine learning remains challenging due to the following facts: 1) the massive samples from the current damage imagery are difficult to be labeled and thus cannot satisfy the training requirement of deep learning, and 2) the similarity between partially damaged and undamaged buildings is high, hindering accurate classification. Leveraging the abundant samples of auxiliary domains, domain adaptation aims to transfer a classifier trained by historical damage imagery to the current task. However, traditional domain adaptation approaches do not fully consider the category-specific information during feature adaptation, which might cause negative transfer. To address this issue, we propose a novel domain adaptation framework that individually aligns each category of the target domain to that of the source domain. Our method combines the variational autoencoder (VAE) and the Gaussian mixture model (GMM). First, the GMM is established to characterize the distribution of the source domain. Then, the VAE is constructed to extract the feature of the target domain. Finally, the Kullback-Leibler (KL) divergence is minimized to force the feature of the target domain to observe the GMM of the source domain. Two damage detection tasks using post-earthquake and post-hurricane imageries are utilized to verify the effectiveness of our method. Experiments show that the proposed method obtains improvements of 4.4% and 9.5%, respectively, compared with the conventional method.
Yosei SHIBATA Nobuki FUKUNAGA Takahiro ISHINABE Hideo FUJIKAKE
For exploration of the functional use of dielectric anisotropy of liquid crystals (LCs), we investigated the dynamic response of molecular alignment in a nematic-phase LC cell with compressive force-induced flow behavior. The results showed that the initial alignment and thickness of the LC layer affect the capacitance of the cell when mechanical pressure is applied.
Longfei CHEN Yuichi NAKAMURA Kazuaki KONDO Dima DAMEN Walterio MAYOL-CUEVAS
We propose a novel framework for integrating beginners' machine operational experiences with those of experts' to obtain a detailed task model. Beginners can provide valuable information for operation guidance and task design; for example, from the operations that are easy or difficult for them, the mistakes they make, and the strategy they tend to choose. However, beginners' experiences often vary widely and are difficult to integrate directly. Thus, we consider an operational experience as a sequence of hand-machine interactions at hotspots. Then, a few experts' experiences and a sufficient number of beginners' experiences are unified using two aggregation steps that align and integrate sequences of interactions. We applied our method to more than 40 experiences of a sewing task. The results demonstrate good potential for modeling and obtaining important properties of the task.
Junyao RAN Youhua FU Hairong WANG Chen LIU
We propose to use clustered interference alignment for the situation where the backhaul link capacity is limited and the base station is cache-enabled given MIMO interference channels, when the number of Tx-Rx pairs exceeds the feasibility constraint of interference alignment. We optimize clustering with the soft cluster size constraint algorithm by adding a cluster size balancing process. In addition, the CSI overhead is quantified as a system performance indicator along with the average throughput. Simulation results show that cluster size balancing algorithm generates clusters that are more balanced as well as attaining higher long-term throughput than the soft cluster size constraint algorithm. The long-term throughput is further improved under high SNR by reallocating the capacity of the backhaul links based on the clustering results.
Shize KANG Lixin JI Zhenglian LI Xindi HAO Yuehang DING
The goal of cross-lingual entity alignment is to match entities from knowledge graph of different languages that represent the same object in the real world. Knowledge graphs of different languages can share the same ontology which we guess may be useful for entity alignment. To verify this idea, we propose a novel embedding model based on TransC. This model first adopts TransC and parameter sharing model to map all the entities and relations in knowledge graphs to a shared low-dimensional semantic space based on a set of aligned entities. Then, the model iteratively uses reinitialization and soft alignment strategy to perform entity alignment. The experimental results show that, compared with the benchmark algorithms, the proposed model can effectively fuse ontology information and achieve relatively better results.
In this paper, hierarchical interference coordination is proposed that suppresses both intra- and inter-cluster interference (ICI) in clustered wireless networks. Assuming transmitters and receivers are equipped with multiple antennas and complete channel state information is shared among all transmitters within the same cluster, interference alignment (IA) is performed that uses nulls to suppress intra-cluster interference. For ICI mitigation, we propose a null-steering precoder designed on the nullspace of a principal eigenvector of the correlated ICI channels, which eliminates a significant amount of ICI power given the exchange of cluster geometry between neighboring clusters. However, as ICI is negligible for the system in which the distance between clusters are large enough, the proposed scheme may not improve the system performance compared with the pure IA scheme that exploits all spatial degrees of freedom (DoF) to increase multiplexing gain without ICI mitigation. For the efficient interference management between intra- and inter-cluster, we analyze the decision criterion that provides an adaptive transmission mode selection between pure IA and proposed ICI reduction in given network environments. Moreover, a low computational complexity based transmission mode switching algorithm is proposed for irregularly distributed networks.
Ahmed M. BENAYA Osamu MUTA Maha ELSABROUTY
Heterogeneous networks (HetNets) technology is expected to be applied in next generation cellular networks to boost system capacity. However, applying HetNets introduces a significant amount of interference among different tiers within the same cell. In this paper, we propose a weighted rank constrained rank minimization (WRCRM) based interference alignment (IA) approach for three-tier HetNets. The concept of RCRM is applied in a different way to deal with the basic characteristic of different tiers: their different interference tolerance. In the proposed WRCRM approach, interference components at different tiers are weighted with different weighting factors (WFs) to reflect their vulnerability to interference. First, we derive an inner and a loose outer bound on the achievable degrees of freedom (DoF) for the three-tier system that is modeled as a three-user mutually interfering broadcast channel (MIBC). Then, the derived bounds along with the well-known IA feasibility conditions are used to show the effectiveness of the proposed WRCRM approach. Results show that there exist WF values that maximize the achievable interference-free dimensions. Moreover, adjusting the required number of DoF according to the derived bounds improves the performance of the WRCRM approach.
Kazumi TAKEMURA Toshiyuki YOSHIDA
This paper proposes a novel Depth From Defocus (DFD) technique based on the property that two images having different focus settings coincide if they are reblurred with the opposite focus setting, which is referred to as the “cross reblurring” property in this paper. Based on the property, the proposed technique estimates the block-wise depth profile for a target object by minimizing the mean squared error between the cross-reblurred images. Unlike existing DFD techniques, the proposed technique is free of lens parameters and independent of point spread function models. A compensation technique for a possible pixel disalignment between images is also proposed to improve the depth estimation accuracy. The experimental results and comparisons with the other DFD techniques show the advantages of our technique.
Conggai LI Feng LIU Shuchao JIANG Yanli XU
Interference alignment (IA) in temporal domain is important in the case of single-antenna vehicle communications. In this paper, perfect cyclic IA based on propagation delay is extended to the K×2 X channels with two receivers and arbitrary transmitters K≥2, which achieves the maximal multiplexing gain by obtaining the theoretical degree of freedom of 2K/(K+1). We deduce the alignment and separability conditions, and propose a general scheme which is flexible in setting the index of time-slot for IA at the receiver side. Furthermore, the feasibility of the proposed scheme in the two-/three- Euclidean space is analyzed and demonstrated.
Feng LIU Shuping WANG Shengming JIANG Yanli XU
For the three-user X channel, its degree of freedom (DoF) 9/5 has been shown achievable theoretically through asymptotic model with infinite resources, which is impractical. In this article, we explore the propagation delay (PD) feature among different links to maximize the achievable DoF with the minimum cost. Since perfect interference alignment (IA) is impossible for 9 messages within 5 time-slots, at least one extra time-slot should be utilized. By the cyclic polynomial approach, we propose a scheme with the maximum achievable DoF of 5/3 for 10 messages within 6 time-slots. Feasibility conditions in the Euclidean space are also deduced, which demonstrates a quite wide range of node arrangements.
Huu-Anh TRAN Heyan HUANG Phuoc TRAN Shumin SHI Huu NGUYEN
Word order is one of the most significant differences between the Chinese and Vietnamese. In the phrase-based statistical machine translation, the reordering model will learn reordering rules from bilingual corpora. If the bilingual corpora are large and good enough, the reordering rules are exact and coverable. However, Chinese-Vietnamese is a low-resource language pair, the extraction of reordering rules is limited. This leads to the quality of reordering in Chinese-Vietnamese machine translation is not high. In this paper, we have combined Chinese dependency relation and Chinese-Vietnamese word alignment results in order to pre-order Chinese word order to be suitable to Vietnamese one. The experimental results show that our methodology has improved the machine translation performance compared to the translation system using only the reordering models of phrase-based statistical machine translation.
Yutaro KUGE Yosei SHIBATA Takahiro ISHINABE Hideo FUJIKAKE
We have proposed a mortar-shaped structure to improve response time and alignment uniformity of twisted vertically aligned (TVA) mode liquid crystal displays (LCDs) for high-contrast reflective color LCDs. From the results of the simulation, we clarified that response time, alignment uniformity and viewing angle range of TVA-mode LCDs were improved by controlling the liquid crystal alignment axis-symmetrically in each pixel.
Weihua LIU Zhenxiang GAO Ying WANG Zhongfang WANG Yongming WANG
For general multiple-input multiple-output (MIMO) interference networks, determining the feasibility conditions of interference alignment (IA) to achieve the maximum degree of freedom (DoF), is tantamount to accessing the maximum spatial resource of MIMO systems. In this paper, from the view of antenna configuration, we first explore the IA feasibility in the K-user MIMO interference channel (IC), G-cell MIMO interference broadcast channel (IBC) and interference multiple access channel (IMAC). We first give the concept of the equalized antenna, and all antenna configurations are divided into two categories, equalized antennas and non-equalized ones. The feasibility conditions of IA system with equalized antennas are derived, and the feasible and infeasible regions are provided. Furthermore, we study the correlations among IC, IBC and IMAC. Interestingly, the G-cell MIMO IBC and IMAC are two special ICs, and a systemic work on IA feasibility for these three interference channels is provided.
Min YUAN Qianjian XING Zhenguo MA Feng YU Yingke XU
In this letter, we present a novel single-precision floating-point multiply-accumulator (FNA-MAC) to achieve lower hardware resource, reduced computing latency and improved computing accuracy for continuous dot product operations. By further fusing the normalization and alignment in the traditional FMA algorithm, the proposed architecture eliminates the first N-1 normalization and rounding operations for an N-point dot product, and preserves the precision of interim results in a significant bit size that is twice of that in the traditional methods. The normalization and rounding of the final result is processed at the cost of consuming an additional multiply-add operation. The simulation results show that the improvement in computational accuracy is significant. Meanwhile, when comparing to a recently published FMA design, the proposed FNA-MAC can reduce the slice look-up table/flip-flop resource and computing latency by a fact of 18%, 33.3%, respectively.