Searcharxiv⌕ Search

arXiv subjects

Zijian Zhang

Publications and source records attributed to Zijian Zhang.

At least 235 records · Page 13Linked to original sources
arXiv2022

Distance-Aware Precoding for Near-Field Capacity Improvement

Extremely large-scale MIMO (XL-MIMO) is a promising technology to improve the capacity for future 6G networks. With a very large number of antennas, the near-field property of XL-MIMO systems becomes dominant. Unlike the classical far-field line-of-sight (LoS) channel with only one available data stream, the significantly increased degrees of freedom (DoFs) are available in the near-field LoS channel. However, limited by the small number of radio frequency (RF) chains, the existing hybrid precoding architecture widely used for 5G is not able to fully exploit the extra DoFs in the near-field region. In this paper, the available DoFs and the capacity of the near-field LoS channel are theoretically analyzed at first. Then, to exploit the near-field effect as a new possibility for capacity improvement, the distance-aware precoding (DAP) scheme is proposed. We develop the DAP architecture, where a dedicated selection circuit is inserted to connect phase shifters and RF chains. Moreover, each RF chain can be flexibly configured to active or inactive according to the distance-related DoFs in the proposed DAP architecture. Based on the developed DAP architecture, a DAP algorithm is proposed to optimize the number of activated RF chains and precoding matrices to match the increased DoFs in the near-field region. Finally, simulation results verify that, the proposed DAP scheme can efficiently utilize the extra DoFs in the near-field region to improve the spectrum efficiency and the energy efficiency as well.

cs.IT↗
arXiv2022

Pattern-Division Multiplexing for Continuous-Aperture MIMO

In recent years, continuous-aperture multiple-input multiple-output (CAP-MIMO) is reinvestigated to achieve improved communication performance with limited antenna apertures. Unlike the classical MIMO composed of discrete antennas, CAP-MIMO has a continuous antenna surface, which is expected to generate any current distribution (i.e., pattern) and induce controllable spatial electromagnetic waves. In this way, the information can be modulated on the electromagnetic waves, which makes it promising to approach the ultimate capacity of finite apertures. The pattern design for CAP-MIMO is the key factor to determine the communication performance, but it has not been well studied in the literature. In this paper, we propose the pattern-division multiplexing to design the patterns for CAP-MIMO. Specifically, we first derive the system model of a typical multi-user CAP-MIMO system, which allows us to formulate the sum-rate maximization problem. Then, we propose a general pattern-division multiplexing technique to transform the design of continuous pattern functions to the design of their projection lengths on finite orthogonal bases. Based on this technique, we further propose a pattern design scheme to solve the formulated sum-rate maximization problem. Simulation results show that, the sum-rate achieved by the proposed scheme is about 260% higher than that achieved by the benchmark scheme.

cs.IT↗
arXiv2021

Finite-Time Capacity: Making Exceed-Shannon Possible?

Shannon-Hartley theorem can accurately calculate the channel capacity when the signal observation time is infinite. However, the calculation of finite-time capacity, which remains unknown, is essential for guiding the design of practical communication systems. In this paper, we investigate the capacity between two correlated Gaussian processes within a finite-time observation window. We first derive the finite-time capacity by providing a limit expression. Then we numerically compute the maximum transmission rate within a single finite-time window. We reveal that the number of bits transmitted per second within the finite-time window can exceed the classical Shannon capacity, which is called as the Exceed-Shannon phenomenon. Furthermore, we derive a finite-time capacity formula under a typical signal autocorrelation case by utilizing the Mercer expansion of trace class operators, and reveal the connection between the finite-time capacity problem and the operator theory. Finally, we analytically prove the existence of the Exceed-Shannon phenomenon in this typical case, and demonstrate the achievability of the finite-time capacity and its compatibility with the classical Shannon capacity.

cs.IT↗
arXiv2021

Position-based Contributive Embeddings for Aspect-Based Sentiment Analysis

Aspect-based sentiment analysis (ABSA), exploring sentiment polarity of aspect-given sentence, is a fine-grained task in the field of nature language processing. Previously researches typically tend to predict polarity based on the meaning of aspect and opinions. However, those approaches mainly focus on considering relations implicitly at the word level, ignore the historical impact of other positional words when the aspect appears in a certain position. Therefore, we propose a Position-based Contributive Embeddings (PosCE) to highlight the historical reference to special position aspect. Contribution of each positional words to the polarity is similar to the process of fairly distributing gains to several actors working in coalition (game theory). Therefore, we quote from the method of Shapley Value and finally gain PosCE to enhance the aspect-based representation for ABSA task. Furthermore, the PosCE can also be used for improving performances on multimodal ABSA task. Extensive experiments on both text and text-audio level using SemEval dataset show that the mainstream models advance performance in accuracy and F1 (increase 2.82% and 4.21% on average respectively) by applying our PosCE.

cs.CL↗
arXiv2021

Compact User-Specific Reconfigurable Intelligent Surfaces for Uplink Transmission

Large-scale antenna arrays employed by the base station (BS) constitute an essential next-generation communications technique. However, due to the constraints of size, cost, and power consumption, it is usually considered unrealistic to use a large-scale antenna array at the user side. Inspired by the emerging technique of reconfigurable intelligent surfaces (RIS), we firstly propose the concept of user-side RIS (US-RIS) for facilitating the employment of a large-scale antenna array at the user side in a cost- and energy-efficient way. In contrast to the existing employments of RIS, which belong to the family of base-station-side RISs (BSS-RISs), the US-RIS concept by definition facilitates the employment of RIS at the user side for the first time. This is achieved by conceiving a multi-layer structure to realize a compact form-factor. Furthermore, our theoretical results demonstrate that, in contrast to the existing single-layer structure, where only the phase of the signal reflected from RIS can be adjusted, the amplitude of the signal penetrating multi-layer US-RIS can also be partially controlled, which brings about a new degree of freedom (DoF) for beamformer design that can be beneficially exploited for performance enhancement. In addition, based on the proposed multi-layer US-RIS, we formulate the signal-to-noise ratio (SNR) maximization problem of US-RIS-aided communications. Due to the non-convexity of the problem introduced by this multi-layer structure, we propose a multi-layer transmit beamformer design relying on an iterative algorithm for finding the optimal solution by alternately updating each variable. Finally, our simulation results verify the superiority of the proposed multi-layer US-RIS as a compact realization of a large-scale antenna array at the user side for uplink transmission.

cs.IT↗
arXiv2021

FaxPlainAC: A Fact-Checking Tool Based on EXPLAINable Models with HumAn Correction in the Loop

Fact-checking on the Web has become the main mechanism through which we detect the credibility of the news or information. Existing fact-checkers verify the authenticity of the information (support or refute the claim) based on secondary sources of information. However, existing approaches do not consider the problem of model updates due to constantly increasing training data due to user feedback. It is therefore important to conduct user studies to correct models' inference biases and improve the model in a life-long learning manner in the future according to the user feedback. In this paper, we present FaxPlainAC, a tool that gathers user feedback on the output of explainable fact-checking models. FaxPlainAC outputs both the model decision, i.e., whether the input fact is true or not, along with the supporting/refuting evidence considered by the model. Additionally, FaxPlainAC allows for accepting user feedback both on the prediction and explanation. Developed in Python, FaxPlainAC is designed as a modular and easily deployable tool. It can be integrated with other downstream tasks and allowing for fact-checking human annotation gathering and life-long learning.

cs.AI↗
arXiv2021

A joint precoding framework for wideband reconfigurable intelligent surface-aided cell-free network

Thanks to the strong ability against the inter-cell interference, cell-free network is considered as a promising technique to improve network capacity. However, further capacity improvement requires to deploy more base stations (BSs) with high cost and power consumption. To address this issue, inspired by the recently developed reconfigurable intelligent surface (RIS) technique, we propose the concept of RIS-aided cell-free network to improve the capacity with low cost and power consumption. The key idea is to replace some of the required BSs by low-cost and energy-efficient RISs. Then, in a wideband RIS-aided cell-free network, we formulate the problem of joint precoding design at BSs and RISs to maximize the network capacity. Due to the non-convexity and high complexity of the formulated problem, we develop an alternating optimization framework to solve this challenging problem. In particular, we decouple this problem via fractional programming, and solve the subproblems alternatively. Note that most of the scenarios considered in existing works are special cases of the general scenario studied in this paper, and the proposed joint precoding framework can serve as a general solution to maximize the capacity in most existing RIS-aided scenarios. Finally, simulation results demonstrate that, compared with the conventional cell-free network, the network capacity under the proposed scheme can be improved significantly.

eess.SP↗
arXiv2021

Learnt Sparsification for Interpretable Graph Neural Networks

Graph neural networks (GNNs) have achieved great success on various tasks and fields that require relational modeling. GNNs aggregate node features using the graph structure as inductive biases resulting in flexible and powerful models. However, GNNs remain hard to interpret as the interplay between node features and graph structure is only implicitly learned. In this paper, we propose a novel method called Kedge for explicitly sparsifying the underlying graph by removing unnecessary neighbors. Our key idea is based on a tractable method for sparsification using the Hard Kumaraswamy distribution that can be used in conjugation with any GNN model. Kedge learns edge masks in a modular fashion trained with any GNN allowing for gradient based optimization in an end-to-end fashion. We demonstrate through extensive experiments that our model Kedge can prune a large proportion of the edges with only a minor effect on the test accuracy. Specifically, in the PubMed dataset, Kedge learns to drop more than 80% of the edges with an accuracy drop of merely 2% showing that graph structure has only a small contribution in comparison to node features. Finally, we also show that Kedge effectively counters the over-smoothing phenomena in deep GNNs by maintaining good task performance with increasing GNN layers.

cs.LG↗
arXiv2021

Explain and Predict, and then Predict Again

A desirable property of learning systems is to be both effective and interpretable. Towards this goal, recent models have been proposed that first generate an extractive explanation from the input text and then generate a prediction on just the explanation called explain-then-predict models. These models primarily consider the task input as a supervision signal in learning an extractive explanation and do not effectively integrate rationales data as an additional inductive bias to improve task performance. We propose a novel yet simple approach ExPred, that uses multi-task learning in the explanation generation phase effectively trading-off explanation and prediction losses. And then we use another prediction network on just the extracted explanations for optimizing the task performance. We conduct an extensive evaluation of our approach on three diverse language datasets -- fact verification, sentiment classification, and QA -- and find that we substantially outperform existing approaches.

cs.CL↗
arXiv2021

Dissonance Between Human and Machine Understanding

Complex machine learning models are deployed in several critical domains including healthcare and autonomous vehicles nowadays, albeit as functional black boxes. Consequently, there has been a recent surge in interpreting decisions of such complex models in order to explain their actions to humans. Models that correspond to human interpretation of a task are more desirable in certain contexts and can help attribute liability, build trust, expose biases and in turn build better models. It is, therefore, crucial to understand how and which models conform to human understanding of tasks. In this paper, we present a large-scale crowdsourcing study that reveals and quantifies the dissonance between human and machine understanding, through the lens of an image classification task. In particular, we seek to answer the following questions: Which (well-performing) complex ML models are closer to humans in their use of features to make accurate predictions? How does task difficulty affect the feature selection capability of machines in comparison to humans? Are humans consistently better at selecting features that make image recognition more accurate? Our findings have important implications on human-machine collaboration, considering that a long term goal in the field of artificial intelligence is to make machines capable of learning and reasoning like humans.

cs.AI↗
arXiv2020

Capacity Improvement in Wideband Reconfigurable Intelligent Surface-Aided Cell-Free Network

Thanks to the strong ability against the inter-cell interference, cell-free network has been considered as a promising technique to improve the network capacity of future wireless systems. However, for further capacity enhancement, it requires to deploy more base stations (BSs) with high cost and power consumption. To address the issue, inspired by the recently proposed technique called reconfigurable intelligent surface (RIS), we propose the concept of RIS-aided cell-free network to improve the network capacity with low cost and power consumption. Then, for the proposed RIS-aided cell-free network in the typical wideband scenario, we formulate the joint precoding design problem at the BSs and RISs to maximize the network capacity. Due to the non-convexity and high complexity of the formulated problem, we develop an alternating optimization algorithm to solve this challenging problem. Note that most of the considered scenarios in existing works are special cases of the general scenario in this paper, and the proposed joint precoding framework can also serve as a general solution to maximize the capacity in most of existing RIS-aided scenarios. Finally, simulation results verify that, compared with the conventional cell-free network, the network capacity of the proposed scheme can be improved significantly.

cs.IT↗
arXiv2020

Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks

Recently, increasing attention has been drawn to the internal mechanisms of convolutional neural networks, and the reason why the network makes specific decisions. In this paper, we develop a novel post-hoc visual explanation method called Score-CAM based on class activation mapping. Unlike previous class activation mapping based approaches, Score-CAM gets rid of the dependence on gradients by obtaining the weight of each activation map through its forward passing score on target class, the final result is obtained by a linear combination of weights and activation maps. We demonstrate that Score-CAM achieves better visual performance and fairness for interpreting the decision making process. Our approach outperforms previous methods on both recognition and localization tasks, it also passes the sanity check. We also indicate its application as debugging tools. Official code has been released.

cs.CV↗
arXiv2020

Convolutional Hierarchical Attention Network for Query-Focused Video Summarization

Previous approaches for video summarization mainly concentrate on finding the most diverse and representative visual contents as video summary without considering the user's preference. This paper addresses the task of query-focused video summarization, which takes user's query and a long video as inputs and aims to generate a query-focused video summary. In this paper, we consider the task as a problem of computing similarity between video shots and query. To this end, we propose a method, named Convolutional Hierarchical Attention Network (CHAN), which consists of two parts: feature encoding network and query-relevance computing module. In the encoding network, we employ a convolutional network with local self-attention mechanism and query-aware global attention mechanism to learns visual information of each shot. The encoded features will be sent to query-relevance computing module to generate queryfocused video summary. Extensive experiments on the benchmark dataset demonstrate the competitive performance and show the effectiveness of our approach.

cs.CV↗
arXiv2020

The propagating mechanism of Chapman-Jouguet deflagration

The deflagration-to-detonation transition (DDT) process is of great importance to both combustion theory and industry safety. In this study, the propagating mechanism of Chapman-Jouguet (C-J) deflagration is studied. Firstly, three models are put forth to decouple the C-J detonation front. These three models are (a) to introduce an expansion parameter into the one-dimensional energy equation, (b) to increase the activation energy of the chemical reaction model and (c) to decouple the shock wave from the flame front by artificial method. The C-J deflagration is obtained after the C-J detonation is decoupled by one-dimensional numerical simulations with different models, chemical reaction kinetics and numerical schemes. Secondly, the propagating mechanism of C-J deflagration is discussed. For the C-J deflagration with a propagating velocity of about 1/2 C-J detonation, the static temperature behind the leading shock wave is too low to ignite the combustion. But, the total temperature of the flow induced by the leading shock wave is high enough to ignite the mixture. The induced flow is slowed down by the rarefaction waves form the wall and its static temperature increases. The flame and the leading shock wave propagate with almost the same velocity and the double-discontinuity structure of the flow field keeps stable. The propagating velocity equals to the sound speed of the combustion products, which is about 1/2 C-J detonation velocity.

physics.flu-dyn↗
arXiv2019

XDeep: An Interpretation Tool for Deep Neural Networks

XDeep is an open-source Python package developed to interpret deep models for both practitioners and researchers. Overall, XDeep takes a trained deep neural network (DNN) as the input, and generates relevant interpretations as the output with the post-hoc manner. From the functionality perspective, XDeep integrates a wide range of interpretation algorithms from the state-of-the-arts, covering different types of methodologies, and is capable of providing both local explanation and global explanation for DNN when interpreting model behaviours. With the well-documented API designed in XDeep, end-users can easily obtain the interpretations for their deep models at hand with several lines of codes, and compare the results among different algorithms. XDeep is generally compatible with Python 3, and can be installed through Python Package Index (PyPI). The source codes are available at: https://github.com/datamllab/xdeep.

cs.LG↗
arXiv2019

Contextual Local Explanation for Black Box Classifiers

We introduce a new model-agnostic explanation technique which explains the prediction of any classifier called CLE. CLE gives an faithful and interpretable explanation to the prediction, by approximating the model locally using an interpretable model. We demonstrate the flexibility of CLE by explaining different models for text, tabular and image classification, and the fidelity of it by doing simulated user experiments.

cs.LG↗
arXiv2019

Analysis and Optimization of Outage Probability in Multi-Intelligent Reflecting Surface-Assisted Systems

Intelligent reflecting surface (IRS) is envisioned to be a promising solution for designing spectral and energy efficient wireless systems. In this letter, we study a multi-IRS-assisted system under Rician fading where the phase shifts adapt to only the line of sight (LoS) components. First, we analyze and optimize the outage probability of the multi-IRS-assisted system in the slow fading scenario for the non-LoS (NLoS) components. We also show that the optimal outage probability decreases with the numbers of IRSs and elements of each IRS when the LoS components are stronger than the NLoS ones. Then, we characterize the asymptotically optimal outage probability in the high signal-to-noise ratio (SNR) regime, and show that it decreases with the powers of the LoS components. To the best of our knowledge, this is the first work that studies the outage probability in multi-IRS-assisted systems.

cs.IT↗