arXiv subjects

Zhi Chen

Publications and source records attributed to Zhi Chen.

At least 235 records · Page 13Linked to original sources

arXiv2022

Beam Training and Alignment for RIS-Assisted Millimeter Wave Systems:State of the Art and Beyond

Reconfigurable intelligent surface (RIS) has recently emerged as a promising paradigm for future cellular networks. Specifically, due to its capability in reshaping the propagation environment, RIS was introduced to address the blockage issue in millimeter Wave (mmWave) or even Terahertz (THz) communications. The deployment of RIS, however, complicates the system architecture and poses a significant challenge for beam training (BT)/ beam alignment (BA), a process that is required to establish a reliable link between the transmitter and the receiver. In this article, we first review several state-of-the-art beam training solutions for RIS-assisted mmWave systems and discuss their respective advantages and limitations. We also present a new multi-directional BT method, which can achieve a decent BA performance with only a small amount of training overhead. Finally, we outline several important open issues in BT for RIS-assisted mmWave systems.

eess.SP↗

arXiv2022

Hyper-relationship Learning Network for Scene Graph Generation

Generating informative scene graphs from images requires integrating and reasoning from various graph components, i.e., objects and relationships. However, current scene graph generation (SGG) methods, including the unbiased SGG methods, still struggle to predict informative relationships due to the lack of 1) high-level inference such as transitive inference between relationships and 2) efficient mechanisms that can incorporate all interactions of graph components. To address the issues mentioned above, we devise a hyper-relationship learning network, termed HLN, for SGG. Specifically, the proposed HLN stems from hypergraphs and two graph attention networks (GATs) are designed to infer relationships: 1) the object-relationship GAT or OR-GAT to explore interactions between objects and relationships, and 2) the hyper-relationship GAT or HR-GAT to integrate transitive inference of hyper-relationships, i.e., the sequential relationships between three objects for transitive reasoning. As a result, HLN significantly improves the performance of scene graph generation by integrating and reasoning from object interactions, relationship interactions, and transitive inference of hyper-relationships. We evaluate HLN on the most popular SGG dataset, i.e., the Visual Genome dataset, and the experimental results demonstrate its great superiority over recent state-of-the-art methods. For example, the proposed HLN improves the recall per relationship from 11.3\% to 13.1\%, and maintains the recall per image from 19.8\% to 34.9\%. We will release the source code and pretrained models on GitHub.

cs.CV↗

arXiv2022

Jigsaw Puzzle: Selective Backdoor Attack to Subvert Malware Classifiers

Malware classifiers are subject to training-time exploitation due to the need to regularly retrain using samples collected from the wild. Recent work has demonstrated the feasibility of backdoor attacks against malware classifiers, and yet the stealthiness of such attacks is not well understood. In this paper, we investigate this phenomenon under the clean-label setting (i.e., attackers do not have complete control over the training or labeling process). Empirically, we show that existing backdoor attacks in malware classifiers are still detectable by recent defenses such as MNTD. To improve stealthiness, we propose a new attack, Jigsaw Puzzle (JP), based on the key observation that malware authors have little to no incentive to protect any other authors' malware but their own. As such, Jigsaw Puzzle learns a trigger to complement the latent patterns of the malware author's samples, and activates the backdoor only when the trigger and the latent pattern are pieced together in a sample. We further focus on realizable triggers in the problem space (e.g., software code) using bytecode gadgets broadly harvested from benign software. Our evaluation confirms that Jigsaw Puzzle is effective as a backdoor, remains stealthy against state-of-the-art defenses, and is a threat in realistic settings that depart from reasoning about feature-space only attacks. We conclude by exploring promising approaches to improve backdoor defenses.

cs.CR↗

arXiv2021

DetarNet: Decoupling Translation and Rotation by Siamese Network for Point Cloud Registration

Point cloud registration is a fundamental step for many tasks. In this paper, we propose a neural network named DetarNet to decouple the translation $t$ and rotation $R$, so as to overcome the performance degradation due to their mutual interference in point cloud registration. First, a Siamese Network based Progressive and Coherent Feature Drift (PCFD) module is proposed to align the source and target points in high-dimensional feature space, and accurately recover translation from the alignment process. Then we propose a Consensus Encoding Unit (CEU) to construct more distinguishable features for a set of putative correspondences. After that, a Spatial and Channel Attention (SCA) block is adopted to build a classification network for finding good correspondences. Finally, the rotation is obtained by Singular Value Decomposition (SVD). In this way, the proposed network decouples the estimation of translation and rotation, resulting in better performance for both of them. Experimental results demonstrate that the proposed DetarNet improves registration performance on both indoor and outdoor scenes. Our code will be available in \url{https://github.com/ZhiChen902/DetarNet}.

cs.CV↗

arXiv2021

Few-Shot NLU with Vector Projection Distance and Abstract Triangular CRF

Data sparsity problem is a key challenge of Natural Language Understanding (NLU), especially for a new target domain. By training an NLU model in source domains and applying the model to an arbitrary target domain directly (even without fine-tuning), few-shot NLU becomes crucial to mitigate the data scarcity issue. In this paper, we propose to improve prototypical networks with vector projection distance and abstract triangular Conditional Random Field (CRF) for the few-shot NLU. The vector projection distance exploits projections of contextual word embeddings on label vectors as word-label similarities, which is equivalent to a normalized linear model. The abstract triangular CRF learns domain-agnostic label transitions for joint intent classification and slot filling tasks. Extensive experiments demonstrate that our proposed methods can significantly surpass strong baselines. Specifically, our approach can achieve a new state-of-the-art on two few-shot NLU benchmarks (Few-Joint and SNIPS) in Chinese and English without fine-tuning on target domains.

cs.CL↗

arXiv2021

Development of a GPU-accelerated Monte Carlo dose calculation module for nuclear medicine, ARCHER-NM: Demonstration for a PET/CT imaging procedure

This paper describes the development and validation of a Monte Carlo (MC) dose computing module dedicated to organ dose calculations of patients undergoing nuclear medicine (NM) internal radiation exposures involving 18F-FDG PET/CT examination. This new module extends the more-than-10-years-long ARCHER project that developed a GPU-accelerated MC dose engine by adding dedicated NM source-definition features. To validate the code, we compared dose distributions from the 0.511-MeV point photon source calculated for a water phantom as well as a patient PET/CT phantom against a well-tested MC code, GATE. The water-phantom results show excellent agreement, suggesting that the radiation physics module in the new NM code is adequate. To demonstrate the clinical utility and advantage of ARCHER-NM, one set of PET/CT data for an adult male NM patient is calculated using the new code. Radiosensitive organs in the CT dataset are segmented using a CNN-based tool called DeepViewer. The PET image intensity maps are converted to radioactivity distributions to allow for MC radiation transport dose calculations at the voxel level. The dose rate maps and corresponding statistical uncertainties were calculated for the duration of PET image acquisition. The dose rate results of the 18F-FDG PET imaging patient show that ARCHER-NM's results agree very well with those of the GATE within 0.58% to 4.11%. Most impressively, ARCHER-NM obtains such results in less than 0.5 minutes while it takes GATE as much as 376 minutes. This is the first study presenting GPU-accelerated patient-specific MC internal radiation dose rate calculations for clinically realistic 18F-FDG PET/CT imaging cases involving auto-segmentation of whole-body PET/CT images. This study suggests that modern computing tools -- ARCHER-NM and DeepViewer -- are accurate and fast enough for routine internal dosimetry in NM clinics.

physics.med-ph↗

arXiv2021

Global Regularity and instability for the incompressible non-viscous Oldroyd-B model

In this paper, we consider the 2-dimensional non-viscous Oldroyd-B model. In the case of the ratio equal 1~($α=0$), it is a difficult case since the velocity field $u(t,x)$ is no longer decay. Fortunately, by {observing the exponential decay} of the stress tensor $τ(t,x)$, we succeeded in proving the global existence for this system with some large initial data. Moreover, we give an unsteady result: when the ratio is close to 1~($a\rightarrow 0$), the system is not steady for large time. This implies an interesting physical phenomenon that the term $a\mathbb{D}u$ is a bridge between the transformation of kinetic energy $u$ and elastic potential energy $τ$, but this process is transient for large time, which leads the instability.

math.AP↗

arXiv2021

Domain Adaptive Semantic Segmentation without Source Data

Domain adaptive semantic segmentation is recognized as a promising technique to alleviate the domain shift between the labeled source domain and the unlabeled target domain in many real-world applications, such as automatic pilot. However, large amounts of source domain data often introduce significant costs in storage and training, and sometimes the source data is inaccessible due to privacy policies. To address these problems, we investigate domain adaptive semantic segmentation without source data, which assumes that the model is pre-trained on the source domain, and then adapting to the target domain without accessing source data anymore. Since there is no supervision from the source domain data, many self-training methods tend to fall into the ``winner-takes-all'' dilemma, where the {\it majority} classes totally dominate the segmentation networks and the networks fail to classify the {\it minority} classes. Consequently, we propose an effective framework for this challenging problem with two components: positive learning and negative learning. In positive learning, we select the class-balanced pseudo-labeled pixels with intra-class threshold, while in negative learning, for each pixel, we investigate which category the pixel does not belong to with the proposed heuristic complementary label selection. Notably, our framework can be easily implemented and incorporated with other methods to further enhance the performance. Extensive experiments on two widely-used synthetic-to-real benchmarks demonstrate our claims and the effectiveness of our framework, which outperforms the baseline with a large margin. Code is available at \url{https://github.com/fumyou13/LDBE}.

cs.CV↗

arXiv2021

Joint Hardware Design and Capacity Analysis for Intelligent Reflecting Surface Enabled Terahertz MIMO Communications

Terahertz (THz) communications have been envisioned as a promising enabler to provide ultra-high data transmission for sixth generation (6G) wireless networks. To tackle the blockage vulnerability brought by severe path attenuation and poor diffraction of THz waves, an intelligent reflecting surface (IRS) is put forward to smartly control the incident THz waves by adjusting the phase shifts. In this paper, we firstly design an efficient hardware structure of graphene-based IRS with phase response up to 306.82 degrees. Subsequently, to characterize the capacity of the IRS-enabled THz multiple-input multiple-output (MIMO) system, an adaptive gradient descent (A-GD) algorithm is developed by dynamically updating the step size during the iterative process, which is determined by the second-order Taylor expansion formulation. In contrast with conventional gradient descent (C-GD) algorithm with fixed step size, the A-GD algorithm evidently improves the achievable rate performance. However, both A-GD algorithm and C-GD algorithm inherit the unacceptable complexity. Then a low complexity alternating optimization (AO) algorithm is proposed by alternately optimizing the precoding matrix by a column-by-column (CBC) algorithm and the phase shift matrix of the IRS by a linear search algorithm. Ultimately, the numerical results demonstrate the effectiveness of the designed hardware structure and the considered algorithms.

cs.IT↗

arXiv2021

Adaptive Feedforward Reference Design for Active Vibration Rejection in Multi-Actuator Hard Disk Drives

In December 2017, Seagate unveiled the Multi Actuator Technology to double the data performance of the future generation hard disk drives (HDD). This technology will equip drives with two dual stage actuators (DSA) each comprising of a voice coil motor (VCM) actuator and a piezoelectric micro actuator (MA) operating on the same pivot point. Each DSA is responsible for controlling half of the drive's arms. As both the DSAs operate independently on the same pivot timber, the control forces and torques generated by one can affect the operation of the other and thereby worsening the performance drastically. In this paper, a robust adaptive feedforward controller is designed as an add-on controller to an existing stabilizing feedback controller to reject the disturbances transferred through the common pivot timber by shaping the references to the VCM actuator and the total output of the dual stage system.

eess.SY↗

arXiv2021

Data-Driven Strictly Positive Real System Identification with prior System Knowledge

Strictly Positive Real (SPR) transfer functions arise in many areas of engineering like passivity theory in circuit analysis and adaptive control to name a few. In many physical systems, it is possible to conclude that the system is Positive Real (PR) or SPR but system identification algorithms might produce estimates which are not SPR. In this paper, an algorithm to approximate frequency response data with SPR transfer functions using Generalized Orthonormal Basis Functions (GOBFs) is presented. Prior knowledge of the system helps us to get approximate pole locations, which can then be used to construct GOBFs. Next, a convex optimization problem will be formulated to obtain an estimate of the SPR transfer function.

eess.SY↗

arXiv2021

A Unified 3D Beam Training and Tracking Procedure for Terahertz Communication

Terahertz (THz) communication is considered as an attractive way to overcome the bandwidth bottleneck and satisfy the ever-increasing capacity demand in the future. Due to the high directivity and propagation loss of THz waves, a massive MIMO system using beamforming is envisioned as a promising technology in THz communication to realize high-gain and directional transmission. However, pilots, which are the fundamentals for many beamforming schemes, are challenging to be accurately detected in the THz band owing to the severe propagation loss. In this paper, a unified 3D beam training and tracking procedure is proposed to effectively realize the beamforming in THz communications, by considering the line-of-sight (LoS) propagation. In particular, a novel quadruple-uniform planar array (QUPA) architecture is analyzed to enlarge the signal coverage, increase the beam gain, and reduce the beam squint loss. Then, a new 3D grid-based (GB) beam training is developed with low complexity, including the design of the 3D codebook and training protocol. Finally, a simple yet effective grid-based hybrid (GBH) beam tracking is investigated to support THz beamforming in an efficient manner. The communication framework based on this procedure can dynamically trigger beam training/tracking depending on the real-time quality of service. Numerical results are presented to demonstrate the superiority of our proposed beam training and tracking over the benchmark methods.

cs.IT↗

arXiv2021

System Identification in Multi-Actuator Hard Disk Drives with Colored Noises using Observer/Kalman Filter Identification (OKID) Framework

Multi Actuator Technology in Hard Disk drives (HDDs) equips drives with two dual stage actuators (DSA) each comprising of a voice coil motor (VCM) actuator and a piezoelectric micro actuator (MA) operating on the same pivot point. Each DSA is responsible for controlling half of the drive's arms. As both the DSAs operate independently on the same pivot timber, the control forces and torques generated by one affect the operation of the other. The feedback controllers might not completely reject these transferred disturbances and a need to design feedforward controllers arises, which require a good model of the disturbance process. The usual system identification techniques produce a biased estimate because of the presence of the runout which is a colored noise. In this paper, we use the OKID framework to estimate this disturbance cross transfer function from the VCM control input of one DSA to the output of the other DSA from the collected time series data corrupted by colored noise.

eess.SY↗

arXiv2021

Semantics Disentangling for Generalized Zero-Shot Learning

Generalized zero-shot learning (GZSL) aims to classify samples under the assumption that some classes are not observable during training. To bridge the gap between the seen and unseen classes, most GZSL methods attempt to associate the visual features of seen classes with attributes or to generate unseen samples directly. Nevertheless, the visual features used in the prior approaches do not necessarily encode semantically related information that the shared attributes refer to, which degrades the model generalization to unseen classes. To address this issue, in this paper, we propose a novel semantics disentangling framework for the generalized zero-shot learning task (SDGZSL), where the visual features of unseen classes are firstly estimated by a conditional VAE and then factorized into semantic-consistent and semantic-unrelated latent vectors. In particular, a total correlation penalty is applied to guarantee the independence between the two factorized representations, and the semantic consistency of which is measured by the derived relation network. Extensive experiments conducted on four GZSL benchmark datasets have evidenced that the semantic-consistent features disentangled by the proposed SDGZSL are more generalizable in tasks of canonical and generalized zero-shot learning. Our source code is available at https://github.com/uqzhichen/SDGZSL.

cs.CV↗

arXiv2021

Intelligent Reflecting Surface Assisted Terahertz Communications toward 6G

Terahertz (THz) communications have emerged as a promising candidate to support the heavy data traffic and exploding network capacity in the future 6G wireless networks. However, THz communications are facing many challenges for practical implementation, such as propagation loss, signal blockage, and hardware cost. In this article, an emerging paradigm of intelligent reflecting surface (IRS) assisted THz communications is analyzed, to address the above issues, by leveraging the joint active and passive beamforming to enhance the communication quality and reduce overheads. Aiming at practical implementation, an overview of the currently available approaches of realizing THz active/passive beam steering at transmitter and IRS is presented. Based on these approaches, a beam training strategy for establishing joint beamforming is then investigated in THz communications. Moreover, various emerging and appealing 6G scenarios that integrate IRS into THz communications are envisioned. Open challenges and future research directions for this new paradigm are finally highlighted.

cs.IT↗

arXiv2021

A Sparsity Algorithm with Applications to Corporate Credit Rating

In Artificial Intelligence, interpreting the results of a Machine Learning technique often termed as a black box is a difficult task. A counterfactual explanation of a particular "black box" attempts to find the smallest change to the input values that modifies the prediction to a particular output, other than the original one. In this work we formulate the problem of finding a counterfactual explanation as an optimization problem. We propose a new "sparsity algorithm" which solves the optimization problem, while also maximizing the sparsity of the counterfactual explanation. We apply the sparsity algorithm to provide a simple suggestion to publicly traded companies in order to improve their credit ratings. We validate the sparsity algorithm with a synthetically generated dataset and we further apply it to quarterly financial statements from companies in financial, healthcare and IT sectors of the US market. We provide evidence that the counterfactual explanation can capture the nature of the real statement features that changed between the current quarter and the following quarter when ratings improved. The empirical results show that the higher the rating of a company the greater the "effort" required to further improve credit rating.

q-fin.RM↗

arXiv2021

CausalRec: Causal Inference for Visual Debiasing in Visually-Aware Recommendation

Visually-aware recommendation on E-commerce platforms aims to leverage visual information of items to predict a user's preference. It is commonly observed that user's attention to visual features does not always reflect the real preference. Although a user may click and view an item in light of a visual satisfaction of their expectations, a real purchase does not always occur due to the unsatisfaction of other essential features (e.g., brand, material, price). We refer to the reason for such a visually related interaction deviating from the real preference as a visual bias. Existing visually-aware models make use of the visual features as a separate collaborative signal similarly to other features to directly predict the user's preference without considering a potential bias, which gives rise to a visually biased recommendation. In this paper, we derive a causal graph to identify and analyze the visual bias of these existing methods. In this causal graph, the visual feature of an item acts as a mediator, which could introduce a spurious relationship between the user and the item. To eliminate this spurious relationship that misleads the prediction of the user's real preference, an intervention and a counterfactual inference are developed over the mediator. Particularly, the Total Indirect Effect is applied for a debiased prediction during the testing phase of the model. This causal inference framework is model agnostic such that it can be integrated into the existing methods. Furthermore, we propose a debiased visually-aware recommender system, denoted as CausalRec to effectively retain the supportive significance of the visual information and remove the visual bias. Extensive experiments are conducted on eight benchmark datasets, which shows the state-of-the-art performance of CausalRec and the efficacy of debiasing.

cs.IR↗

arXiv2021

Interpretable Machine Learning: Fundamental Principles and 10 Grand Challenges

Interpretability in machine learning (ML) is crucial for high stakes decisions and troubleshooting. In this work, we provide fundamental principles for interpretable ML, and dispel common misunderstandings that dilute the importance of this crucial topic. We also identify 10 technical challenge areas in interpretable machine learning and provide history and background on each problem. Some of these problems are classically important, and some are recent problems that have arisen in the last few years. These problems are: (1) Optimizing sparse logical models such as decision trees; (2) Optimization of scoring systems; (3) Placing constraints into generalized additive models to encourage sparsity and better interpretability; (4) Modern case-based reasoning, including neural networks and matching for causal inference; (5) Complete supervised disentanglement of neural networks; (6) Complete or even partial unsupervised disentanglement of neural networks; (7) Dimensionality reduction for data visualization; (8) Machine learning models that can incorporate physics and other generative or causal constraints; (9) Characterization of the "Rashomon set" of good models; and (10) Interpretable reinforcement learning. This survey is suitable as a starting point for statisticians and computer scientists interested in working in interpretable machine learning.

cs.LG↗