Showing 1–50 of 78 results for author: Hui, S

A new platooning model for connected and autonomous vehicles to improve string stability

Authors: Shouwei Hui, Michael Zhang

Abstract: This paper introduces a novel idea of coordinated vehicle platooning such that platoon followers inside the platoon communicates only to the platoon leader. A novel dynamic model is proposed to take driving safety into account when there is communication delay. Some general results of linear stability are proved mathematically, and numerical simulations are conducted to show the effect of model pa… ▽ More This paper introduces a novel idea of coordinated vehicle platooning such that platoon followers inside the platoon communicates only to the platoon leader. A novel dynamic model is proposed to take driving safety into account when there is communication delay. Some general results of linear stability are proved mathematically, and numerical simulations are conducted to show the effect of model parameters for both ring road with an initial disturbance and infinite road with a periodic disturbance. The simulation results are consistent with theoretical analysis, and demonstrate that the proposed look-to-the-leader platooning strategy is far superior than the follow-one-vehicle-ahead or follow-two-vehicle-ahead conventional car-following (CF) strategies in stabilizing traffic flow. This paper provides a new perspective for the organization of platoons of autonomous vehicles. △ Less

Submitted 29 May, 2024; originally announced May 2024.

Comments: preprint submitted to Physica A

Exploring Automated Contouring Across Institutional Boundaries: A Deep Learning Approach with Mouse Micro-CT Datasets

Authors: Lu Jiang, Di Xu, Qifan Xu, Arion Chatziioannou, Keisuke S. Iwamoto, Susanta Hui, Ke Sheng

Abstract: Image-guided mouse irradiation is essential to understand interventions involving radiation prior to human studies. Our objective is to employ Swin UNEt Transformers (Swin UNETR) to segment native micro-CT and contrast-enhanced micro-CT scans and benchmark the results against 3D no-new-Net (nnU-Net). Swin UNETR reformulates mouse organ segmentation as a sequence-to-sequence prediction task, using… ▽ More Image-guided mouse irradiation is essential to understand interventions involving radiation prior to human studies. Our objective is to employ Swin UNEt Transformers (Swin UNETR) to segment native micro-CT and contrast-enhanced micro-CT scans and benchmark the results against 3D no-new-Net (nnU-Net). Swin UNETR reformulates mouse organ segmentation as a sequence-to-sequence prediction task, using a hierarchical Swin Transformer encoder to extract features at 5 resolution levels, and connects to a Fully Convolutional Neural Network (FCNN)-based decoder via skip connections. The models were trained and evaluated on open datasets, with data separation based on individual mice. Further evaluation on an external mouse dataset acquired on a different micro-CT with lower kVp and higher imaging noise was also employed to assess model robustness and generalizability. Results indicate that Swin UNETR consistently outperforms nnU-Net and AIMOS in terms of average dice similarity coefficient (DSC) and Hausdorff distance (HD95p), except in two mice of intestine contouring. This superior performance is especially evident in the external dataset, confirming the model's robustness to variations in imaging conditions, including noise and quality, thereby positioning Swin UNETR as a highly generalizable and efficient tool for automated contouring in pre-clinical workflows. △ Less

Submitted 28 May, 2024; originally announced May 2024.

Expert Insight-Enhanced Follow-up Chest X-Ray Summary Generation

Authors: Zhichuan Wang, Kinhei Lee, Qiao Deng, Tiffany Y. So, Wan Hang Chiu, Yeung Yu Hui, Bing**g Zhou, Edward S. Hui

Abstract: A chest X-ray radiology report describes abnormal findings not only from X-ray obtained at current examination, but also findings on disease progression or change in device placement with reference to the X-ray from previous examination. Majority of the efforts on automatic generation of radiology report pertain to reporting the former, but not the latter, type of findings. To the best of the auth… ▽ More A chest X-ray radiology report describes abnormal findings not only from X-ray obtained at current examination, but also findings on disease progression or change in device placement with reference to the X-ray from previous examination. Majority of the efforts on automatic generation of radiology report pertain to reporting the former, but not the latter, type of findings. To the best of the authors' knowledge, there is only one work dedicated to generating summary of the latter findings, i.e., follow-up summary. In this study, we therefore propose a transformer-based framework to tackle this task. Motivated by our observations on the significance of medical lexicon on the fidelity of summary generation, we introduce two mechanisms to bestow expert insight to our model, namely expert soft guidance and masked entity modeling loss. The former mechanism employs a pretrained expert disease classifier to guide the presence level of specific abnormalities, while the latter directs the model's attention toward medical lexicon. Extensive experiments were conducted to demonstrate that the performance of our model is competitive with or exceeds the state-of-the-art. △ Less

Submitted 6 May, 2024; v1 submitted 1 May, 2024; originally announced May 2024.

Comments: accepted by 22nd International Conference on Artificial Intelligence in medicine (AIME2024)

ACM Class: I.2.1

Gradient estimation of a generalized non-linear heat type equation along Super-Perelman Ricci flow on weighted Riemannian manifolds

Authors: Suraj Ghosh, Abimbola Abolarinwa, Shyamal Kumar Hui

Abstract: In this article we derive gradient estimation for positive solution of the equation \begin{equation*} (\partial_t-Δ_f)u = A(u)p(x,t) + B(u)q(x,t) + \mathcal{G}(u) \end{equation*} on a weighted Riemannian manifold evolving along the $(k,m)$ super Perelman-Ricci flow \begin{equation*} \frac{\partial g}{\partial t}(x,t)+2Ric_f^m(g)(x,t)\ge -2kg(x,t). \end{equation*} As an application of gradient esti… ▽ More In this article we derive gradient estimation for positive solution of the equation \begin{equation*} (\partial_t-Δ_f)u = A(u)p(x,t) + B(u)q(x,t) + \mathcal{G}(u) \end{equation*} on a weighted Riemannian manifold evolving along the $(k,m)$ super Perelman-Ricci flow \begin{equation*} \frac{\partial g}{\partial t}(x,t)+2Ric_f^m(g)(x,t)\ge -2kg(x,t). \end{equation*} As an application of gradient estimation we derive a Harnack type inequality along with a Liouville type theorem. △ Less

Submitted 24 April, 2024; originally announced April 2024.

MSC Class: 53C21; 58J60; 58J35

Grounded Knowledge-Enhanced Medical VLP for Chest X-Ray

Authors: Qiao Deng, Zhongzhen Huang, Yunqi Wang, Zhichuan Wang, Zhao Wang, Xiaofan Zhang, Qi Dou, Yeung Yu Hui, Edward S. Hui

Abstract: Medical vision-language pre-training has emerged as a promising approach for learning domain-general representations of medical image and text. Current algorithms that exploit the global and local alignment between medical image and text could however be marred by the redundant information in medical data. To address this issue, we propose a grounded knowledge-enhanced medical vision-language pre-… ▽ More Medical vision-language pre-training has emerged as a promising approach for learning domain-general representations of medical image and text. Current algorithms that exploit the global and local alignment between medical image and text could however be marred by the redundant information in medical data. To address this issue, we propose a grounded knowledge-enhanced medical vision-language pre-training (GK-MVLP) framework for chest X-ray. In this framework, medical knowledge is grounded to the appropriate anatomical regions by using a transformer-based grounded knowledge-enhanced module for fine-grained alignment between anatomical region-level visual features and the textural features of medical knowledge. The performance of GK-MVLP is competitive with or exceeds the state of the art on downstream chest X-ray disease classification, disease localization, report generation, and medical visual question-answering tasks. Our results show the advantage of incorporating grounding mechanism to remove biases and improve the alignment between chest X-ray image and radiology report. △ Less

Submitted 23 April, 2024; originally announced April 2024.

Image2Sentence based Asymmetrical Zero-shot Composed Image Retrieval

Authors: Yongchao Du, Min Wang, Wengang Zhou, Shu** Hui, Houqiang Li

Abstract: The task of composed image retrieval (CIR) aims to retrieve images based on the query image and the text describing the users' intent. Existing methods have made great progress with the advanced large vision-language (VL) model in CIR task, however, they generally suffer from two main issues: lack of labeled triplets for model training and difficulty of deployment on resource-restricted environmen… ▽ More The task of composed image retrieval (CIR) aims to retrieve images based on the query image and the text describing the users' intent. Existing methods have made great progress with the advanced large vision-language (VL) model in CIR task, however, they generally suffer from two main issues: lack of labeled triplets for model training and difficulty of deployment on resource-restricted environments when deploying the large vision-language model. To tackle the above problems, we propose Image2Sentence based Asymmetric zero-shot composed image retrieval (ISA), which takes advantage of the VL model and only relies on unlabeled images for composition learning. In the framework, we propose a new adaptive token learner that maps an image to a sentence in the word embedding space of VL model. The sentence adaptively captures discriminative visual information and is further integrated with the text modifier. An asymmetric structure is devised for flexible deployment, in which the lightweight model is adopted for the query side while the large VL model is deployed on the gallery side. The global contrastive distillation and the local alignment regularization are adopted for the alignment between the light model and the VL model for CIR task. Our experiments demonstrate that the proposed ISA could better cope with the real retrieval scenarios and further improve retrieval accuracy and efficiency. △ Less

Submitted 3 March, 2024; originally announced March 2024.

Comments: ICLR 2024 spotlight

Symmetry and pseudosymmetry properties of Vaidya-Bonner-de Sitter spacetime

Authors: Absos Ali Shaikh, Shyamal kumar Hui, Mousumi Sarkar, V. Amarendra Babu

Abstract: The primary focus of the current study is to explore the geometrical properties of the Vaidya-Bonner-de Sitter (briefly, VBdS) spacetime, which is a generalization of Vaidya-Bonner spacetime, Vaidya spacetime and Schwarzschild spacetime. In this study we have shown that the VBdS spacetime describes various types of pseudosymmetric structures, including pseudosymmetry due to conformal curvature, co… ▽ More The primary focus of the current study is to explore the geometrical properties of the Vaidya-Bonner-de Sitter (briefly, VBdS) spacetime, which is a generalization of Vaidya-Bonner spacetime, Vaidya spacetime and Schwarzschild spacetime. In this study we have shown that the VBdS spacetime describes various types of pseudosymmetric structures, including pseudosymmetry due to conformal curvature, conharmonic curvature and other curvatures. Additionally, it is shown that such a spacetime is 2-quasi-Einstein, Einstein manifold of level 3, generalized Roter type, and that conformal 2-forms are recurrent. The geometric features of the Vaidya-Bonner spacetime, Vaidya spacetime, and Schwarzschild spacetime are obtained as a particular instance of the main determination. It is further established that the VBdS spacetime admits almost Ricci soliton and almost η-Yamabe soliton with respect to non-Killing vector fields. Also, it is proved that such a spacetime possesses generalized conharmonic curvature inheritance. It is interesting to note that in the VBdS spacetime the tensors Q(T,R), Q(S,R) and Q(g,R) are linearly dependent. Finally, this spacetime is compared with the Vaidya-Bonner spacetime with respect to their admitting geometric structures, viz., various kinds of symmetry and pseudosymmetry properties. △ Less

Submitted 19 January, 2024; originally announced February 2024.

Comments: 31 pages

MSC Class: 53B20; 53B30; 53B50; 53C15; 53C25; 53C35; 83C15

Understanding the Overfitting of the Episodic Meta-training

Authors: Siqi Hui, San** Zhou, Ye deng, **jun Wang

Abstract: Despite the success of two-stage few-shot classification methods, in the episodic meta-training stage, the model suffers severe overfitting. We hypothesize that it is caused by over-discrimination, i.e., the model learns to over-rely on the superficial features that fit for base class discrimination while suppressing the novel class generalization. To penalize over-discrimination, we introduce kno… ▽ More Despite the success of two-stage few-shot classification methods, in the episodic meta-training stage, the model suffers severe overfitting. We hypothesize that it is caused by over-discrimination, i.e., the model learns to over-rely on the superficial features that fit for base class discrimination while suppressing the novel class generalization. To penalize over-discrimination, we introduce knowledge distillation techniques to keep novel generalization knowledge from the teacher model during training. Specifically, we select the teacher model as the one with the best validation accuracy during meta-training and restrict the symmetric Kullback-Leibler (SKL) divergence between the output distribution of the linear classifier of the teacher model and that of the student model. This simple approach outperforms the standard meta-training process. We further propose the Nearest Neighbor Symmetric Kullback-Leibler (NNSKL) divergence for meta-training to push the limits of knowledge distillation techniques. NNSKL takes few-shot tasks as input and penalizes the output of the nearest neighbor classifier, which possesses an impact on the relationships between query embedding and support centers. By combining SKL and NNSKL in meta-training, the model achieves even better performance and surpasses state-of-the-art results on several benchmarks. △ Less

Submitted 6 July, 2023; v1 submitted 29 June, 2023; originally announced June 2023.

Echoes from charged black holes influenced by quintessence

Authors: Siyuan Hui, Benrong Mu

Abstract: In this paper, we investigate the effective potential and echoes from the dyonic black hole with quintessence. For a dyonic black hole, the quasi-topological electromagnetism provides proper matter energy-momentum tensor to curve the spacetime, and quintessence strengthens this force. We find that when the effect of quintessence becomes stronger, the black hole potential transforms between single-… ▽ More In this paper, we investigate the effective potential and echoes from the dyonic black hole with quintessence. For a dyonic black hole, the quasi-topological electromagnetism provides proper matter energy-momentum tensor to curve the spacetime, and quintessence strengthens this force. We find that when the effect of quintessence becomes stronger, the black hole potential transforms between single-peak and double-peak, which will influence the existence of black hole echoes. In particular, we find that observer will receive a sudden vanishment of high-frequency echoes when quintessence remains a relatively strong effect. △ Less

Submitted 18 May, 2023; originally announced May 2023.

Comments: 15 pages, 8 figures

T-former: An Efficient Transformer for Image Inpainting

Authors: Ye Deng, Siqi Hui, San** Zhou, Deyu Meng, **jun Wang

Abstract: Benefiting from powerful convolutional neural networks (CNNs), learning-based image inpainting methods have made significant breakthroughs over the years. However, some nature of CNNs (e.g. local prior, spatially shared parameters) limit the performance in the face of broken images with diverse and complex forms. Recently, a class of attention-based network architectures, called transformer, has s… ▽ More Benefiting from powerful convolutional neural networks (CNNs), learning-based image inpainting methods have made significant breakthroughs over the years. However, some nature of CNNs (e.g. local prior, spatially shared parameters) limit the performance in the face of broken images with diverse and complex forms. Recently, a class of attention-based network architectures, called transformer, has shown significant performance on natural language processing fields and high-level vision tasks. Compared with CNNs, attention operators are better at long-range modeling and have dynamic weights, but their computational complexity is quadratic in spatial resolution, and thus less suitable for applications involving higher resolution images, such as image inpainting. In this paper, we design a novel attention linearly related to the resolution according to Taylor expansion. And based on this attention, a network called $T$-former is designed for image inpainting. Experiments on several benchmark datasets demonstrate that our proposed method achieves state-of-the-art accuracy while maintaining a relatively low number of parameters and computational complexity. The code can be found at \href{https://github.com/dengyecode/T-former_image_inpainting}{github.com/dengyecode/T-former\_image\_inpainting} △ Less

Submitted 18 May, 2023; v1 submitted 12 May, 2023; originally announced May 2023.

Journal ref: ACM Multimedia 2022

Li-Yau, Hamilton gradient and Hessian estimates for nonlinear weighted parabolic equations and applications

Authors: Shyamal Kumar Hui, Abimbola Abolarinwa, Sujit Bhattacharyya

Abstract: This article is devoted to the study of several estimations for a positive solution to a nonlinear weighted parabolic equation on a weighted Riemannian manifold. We therefore derive new Li-Yau type and Hamilton type gradient estimates yielding several consequences. We also derive Hessian estimate and some corollaries for the same equation. Among the applications of our estimates discussed here are… ▽ More This article is devoted to the study of several estimations for a positive solution to a nonlinear weighted parabolic equation on a weighted Riemannian manifold. We therefore derive new Li-Yau type and Hamilton type gradient estimates yielding several consequences. We also derive Hessian estimate and some corollaries for the same equation. Among the applications of our estimates discussed here are Harnack type inequalities, Liouville type theorems and a local time reversed Harnack inequality. △ Less

Submitted 24 March, 2023; originally announced March 2023.

MSC Class: 53C21; 58J35; 35B45

A Multimodal Data-driven Framework for Anxiety Screening

Authors: Haimiao Mo, Shuai Ding, Siu Cheung Hui

Abstract: Early screening for anxiety and appropriate interventions are essential to reduce the incidence of self-harm and suicide in patients. Due to limited medical resources, traditional methods that overly rely on physician expertise and specialized equipment cannot simultaneously meet the needs for high accuracy and model interpretability. Multimodal data can provide more objective evidence for anxiety… ▽ More Early screening for anxiety and appropriate interventions are essential to reduce the incidence of self-harm and suicide in patients. Due to limited medical resources, traditional methods that overly rely on physician expertise and specialized equipment cannot simultaneously meet the needs for high accuracy and model interpretability. Multimodal data can provide more objective evidence for anxiety screening to improve the accuracy of models. The large amount of noise in multimodal data and the unbalanced nature of the data make the model prone to overfitting. However, it is a non-differentiable problem when high-dimensional and multimodal feature combinations are used as model inputs and incorporated into model training. This causes existing anxiety screening methods based on machine learning and deep learning to be inapplicable. Therefore, we propose a multimodal data-driven anxiety screening framework, namely MMD-AS, and conduct experiments on the collected health data of over 200 seafarers by smartphones. The proposed framework's feature extraction, dimension reduction, feature selection, and anxiety inference are jointly trained to improve the model's performance. In the feature selection step, a feature selection method based on the Improved Fireworks Algorithm is used to solve the non-differentiable problem of feature combination to remove redundant features and search for the ideal feature subset. The experimental results show that our framework outperforms the comparison methods. △ Less

Submitted 15 March, 2023; originally announced March 2023.

On curvature related geometric properties of Hayward black hole spacetime

Authors: Absos Ali Shaikh, Shyamal Kumar Hui, Biswa Ranjan Datta, Mousumi Sarkar

Abstract: This paper is devoted to the study of curvature properties of Hayward black hole (briefly, HBH) spacetime, which is a solution of Einstein field equations (briefly, EFE) having non-vanishing cosmological constant. We have proved that the HBH spacetime is an Einstein manifold of level $2$, $2$-quasi Einstein, generalized quasi-Einstein and Roter type manifold. Also, it is shown that the nature of t… ▽ More This paper is devoted to the study of curvature properties of Hayward black hole (briefly, HBH) spacetime, which is a solution of Einstein field equations (briefly, EFE) having non-vanishing cosmological constant. We have proved that the HBH spacetime is an Einstein manifold of level $2$, $2$-quasi Einstein, generalized quasi-Einstein and Roter type manifold. Also, it is shown that the nature of the HBH spacetime is pseudosymmetric and it obeys several types of pseudosymmetries, such as, pseudosymmetry due to concircular, conformal and conharmonic curvature (i.e., $F\cdot F=\mathcal{L}Q(g,F)$ for $F=W,C, K$ with a smooth scalar function $ \mathcal{L} $), and it also possesses the relation $R\cdot R-\mathcal{L} Q(g,C)=Q(S,R)$. It is engrossing to mention that the nature of energy momentum tensor of the HBH spacetime is pseudosymmetric. On the basis of curvature related properties, we have made a comparison among Reissner-Nordström spacetime, interior black hole spacetime and HBH spacetime. Also, it is shown that the HBH spacetime admits an almost $η$-Ricci soliton as well as an almost $η$-Ricci-Yamabe soliton. Finally, an elegant comparative study is delineated between the HBH spacetime and the point-like global monopole spacetime with respect to different kinds of symmetry, such as, motion, curvature collineation, curvature inheritance etc. △ Less

Submitted 23 February, 2023; originally announced March 2023.

MSC Class: 53B20; 53B30; 53B50; 53C15; 53C25; 53C35; 83C15

Dialogue State Distillation Network with Inter-slot Contrastive Learning for Dialogue State Tracking

Authors: **g Xu, Dandan Song, Chong Liu, Siu Cheung Hui, Fei Li, Qiang Ju, Xiaonan He, Jian Xie

Abstract: In task-oriented dialogue systems, Dialogue State Tracking (DST) aims to extract users' intentions from the dialogue history. Currently, most existing approaches suffer from error propagation and are unable to dynamically select relevant information when utilizing previous dialogue states. Moreover, the relations between the updates of different slots provide vital clues for DST. However, the exis… ▽ More In task-oriented dialogue systems, Dialogue State Tracking (DST) aims to extract users' intentions from the dialogue history. Currently, most existing approaches suffer from error propagation and are unable to dynamically select relevant information when utilizing previous dialogue states. Moreover, the relations between the updates of different slots provide vital clues for DST. However, the existing approaches rely only on predefined graphs to indirectly capture the relations. In this paper, we propose a Dialogue State Distillation Network (DSDN) to utilize relevant information of previous dialogue states and migrate the gap of utilization between training and testing. Thus, it can dynamically exploit previous dialogue states and avoid introducing error propagation simultaneously. Further, we propose an inter-slot contrastive learning loss to effectively capture the slot co-update relations from dialogue context. Experiments are conducted on the widely used MultiWOZ 2.0 and MultiWOZ 2.1 datasets. The experimental results show that our proposed model achieves the state-of-the-art performance for DST. △ Less

Submitted 7 March, 2023; v1 submitted 16 February, 2023; originally announced February 2023.

Comments: Accepted by AAAI 2023

Tensor Robust PCA with Nonconvex and Nonlocal Regularization

Authors: Xiaoyu Geng, Qiang Guo, Shuaixiong Hui, Ming Yang, Caiming Zhang

Abstract: Tensor robust principal component analysis (TRPCA) is a classical way for low-rank tensor recovery, which minimizes the convex surrogate of tensor rank by shrinking each tensor singular value equally. However, for real-world visual data, large singular values represent more significant information than small singular values. In this paper, we propose a nonconvex TRPCA (N-TRPCA) model based on the… ▽ More Tensor robust principal component analysis (TRPCA) is a classical way for low-rank tensor recovery, which minimizes the convex surrogate of tensor rank by shrinking each tensor singular value equally. However, for real-world visual data, large singular values represent more significant information than small singular values. In this paper, we propose a nonconvex TRPCA (N-TRPCA) model based on the tensor adjustable logarithmic norm. Unlike TRPCA, our N-TRPCA can adaptively shrink small singular values more and shrink large singular values less. In addition, TRPCA assumes that the whole data tensor is of low rank. This assumption is hardly satisfied in practice for natural visual data, restricting the capability of TRPCA to recover the edges and texture details from noisy images and videos. To this end, we integrate nonlocal self-similarity into N-TRPCA, and further develop a nonconvex and nonlocal TRPCA (NN-TRPCA) model. Specifically, similar nonlocal patches are grouped as a tensor and then each group tensor is recovered by our N-TRPCA. Since the patches in one group are highly correlated, all group tensors have strong low-rank property, leading to an improvement of recovery performance. Experimental results demonstrate that the proposed NN-TRPCA outperforms existing TRPCA methods in visual data recovery. The demo code is available at https://github.com/qguo2010/NN-TRPCA. △ Less

Submitted 7 July, 2023; v1 submitted 4 November, 2022; originally announced November 2022.

Comments: 15 pages, 7 figures. Submitted to CVIU

Semantic Pivoting Model for Effective Event Detection

Authors: Anran Hao, Siu Cheung Hui, Jian Su

Abstract: Event Detection, which aims to identify and classify mentions of event instances from unstructured articles, is an important task in Natural Language Processing (NLP). Existing techniques for event detection only use homogeneous one-hot vectors to represent the event type classes, ignoring the fact that the semantic meaning of the types is important to the task. Such an approach is inefficient and… ▽ More Event Detection, which aims to identify and classify mentions of event instances from unstructured articles, is an important task in Natural Language Processing (NLP). Existing techniques for event detection only use homogeneous one-hot vectors to represent the event type classes, ignoring the fact that the semantic meaning of the types is important to the task. Such an approach is inefficient and prone to overfitting. In this paper, we propose a Semantic Pivoting Model for Effective Event Detection (SPEED), which explicitly incorporates prior information during training and captures semantically meaningful correlations between input and events. Experimental results show that our proposed model achieves state-of-the-art performance and outperforms the baselines in multiple settings without using any external resources. △ Less

Submitted 1 November, 2022; originally announced November 2022.

Comments: 11 pages, 4 figures; Accepted to ACIIDS 2022

Hamilton and Souplet-Zhang type estimations on semilinear parabolic system along geometric flow

Authors: Shyamal Kumar Hui, Shahroud Azami, Sujit Bhattacharyya

Abstract: In this article we derive both Hamilton type and Souplet-Zhang type gradient estimations for a system of semilinear equations along a geometric flow on a weighted Riemannian manifold. In this article we derive both Hamilton type and Souplet-Zhang type gradient estimations for a system of semilinear equations along a geometric flow on a weighted Riemannian manifold. △ Less

Submitted 26 August, 2022; originally announced August 2022.

Comments: 13 pages

MSC Class: 53C21; 58J60; 58J35; 35B45

Curvature properties of Bardeen black hole spacetime

Authors: Absos Ali Shaikh, Shyamal Kumar Hui, Mousumi Sarkar

Abstract: The Bardeen solution corresponding to Einstein field equations with a cosmological constant is a regular black hole. The main goal of this manuscript is to investigate the geometric structures in terms of curvature conditions admitted by this spacetime. It is found that this spacetime is pseudosymmetric and possess several kinds of pseudosymmetries. Also, it is a manifold of pseudosymmetry Weyl cu… ▽ More The Bardeen solution corresponding to Einstein field equations with a cosmological constant is a regular black hole. The main goal of this manuscript is to investigate the geometric structures in terms of curvature conditions admitted by this spacetime. It is found that this spacetime is pseudosymmetric and possess several kinds of pseudosymmetries. Also, it is a manifold of pseudosymmetry Weyl curvature and the difference tensor C.R-R.C linearly depends on the tensors Q(g;C) and Q(S;C). It is interesting to note that such a spacetime is weakly generalized recurrent manifold and satisfies special recurrent like structure. Further, it is an Einstein manifold of level 2 and Roter type. The energy momentum tensor of this spacetime is pseudosymmetric and finally a worthy comparison between the geometric properties of Bardeen spacetime and Reissner-Nordström spacetime is given. △ Less

Submitted 8 June, 2022; originally announced July 2022.

Horizontal Federated Learning and Secure Distributed Training for Recommendation System with Intel SGX

Authors: Siyuan Hui, Yuqiu Zhang, Albert Hu, Edmund Song

Abstract: With the advent of big data era and the development of artificial intelligence and other technologies, data security and privacy protection have become more important. Recommendation systems have many applications in our society, but the model construction of recommendation systems is often inseparable from users' data. Especially for deep learning-based recommendation systems, due to the complexi… ▽ More With the advent of big data era and the development of artificial intelligence and other technologies, data security and privacy protection have become more important. Recommendation systems have many applications in our society, but the model construction of recommendation systems is often inseparable from users' data. Especially for deep learning-based recommendation systems, due to the complexity of the model and the characteristics of deep learning itself, its training process not only requires long training time and abundant computational resources but also needs to use a large amount of user data, which poses a considerable challenge in terms of data security and privacy protection. How to train a distributed recommendation system while ensuring data security has become an urgent problem to be solved. In this paper, we implement two schemes, Horizontal Federated Learning and Secure Distributed Training, based on Intel SGX(Software Guard Extensions), an implementation of a trusted execution environment, and TensorFlow framework, to achieve secure, distributed recommendation system-based learning schemes in different scenarios. We experiment on the classical Deep Learning Recommendation Model (DLRM), which is a neural network-based machine learning model designed for personalization and recommendation, and the results show that our implementation introduces approximately no loss in model performance. The training speed is within acceptable limits. △ Less

Submitted 11 July, 2022; originally announced July 2022.

Comments: 5 pages, 8 figures

Joule-Thomson expansion of the lower-dimensional black hole in rainbow gravity

Authors: Siyuan Hui, Benrong Mu, Jun Tao

Abstract: In this paper, we extend Joule-Thomson expansion to the low-dimensional regime in rainbow gravity by considering the rainbow rotating BTZ metric in the (2+1)-dimensional spacetime. After the metric of the black hole is obtained, we get the Joule-Thomson expansion of the black hole, including the Joule-Thomson coefficient, inversion curves, and isenthalpic curves. We find that a rainbow rotating BT… ▽ More In this paper, we extend Joule-Thomson expansion to the low-dimensional regime in rainbow gravity by considering the rainbow rotating BTZ metric in the (2+1)-dimensional spacetime. After the metric of the black hole is obtained, we get the Joule-Thomson expansion of the black hole, including the Joule-Thomson coefficient, inversion curves, and isenthalpic curves. We find that a rainbow rotating BTZ black hole does not have $P-V$ critical behavior. The effects of rainbow gravity are to slow down the trend of the increase of the Joule-Thomson coefficient and make its zero point larger. Moreover, the rainbow gravity slows down the inverse temperature of the black hole, meaning that a rainbow rotating BTZ black hole tends to change its heating or cooling action at a lower temperature, which can be attributed to the topology of the black hole. △ Less

Submitted 4 July, 2022; originally announced July 2022.

Comments: 14 pages, 7 figures, 2 tables

Stationary and Free-fall frame Kerr black hole in gravity's rainbow

Authors: Yuzhou Tao, Benrong Mu, Siyuan Hui, Jun Tao

Abstract: Doubly special relativity (DSR) is an effective model for encoding quantum gravity in flat spacetime. To incorporate DSR into general relativity, one could use gravity's rainbow, where the spacetime background felt by a test particle would depend on its energy. In this paper, we investigate the thermodynamics of rainbow Kerr black hole in the scenario with the stationary(ST) orthonormal frame and… ▽ More Doubly special relativity (DSR) is an effective model for encoding quantum gravity in flat spacetime. To incorporate DSR into general relativity, one could use gravity's rainbow, where the spacetime background felt by a test particle would depend on its energy. In this paper, we investigate the thermodynamics of rainbow Kerr black hole in the scenario with the stationary(ST) orthonormal frame and free-fall(FF) orthonormal frame. After the rainbow metric in ST frame and FF frame is deduced, the Hamilton-Jacobi method is used to acquire the modified Hawking temperature, specific heat and corresponding the modified entropy to each scenario, then the thermodynamic properties are discussed. We find that the effects of rainbow gravity on Kerr black holes are quite model-dependent. In other words, the value of parameter $η$ and $n$ with Amelino-Camelia's proposal are crucially important and worth discussing. Specificly, with most widly accepted choice ($n=2,η>0$), the effects of rainbow gravity tend to decrease the Hawking temperature but increase the black hole entropy in ST frame, and increase the Hawking temperature but decrease the black hole entropy in FF frame conversely. △ Less

Submitted 26 May, 2022; originally announced May 2022.

Comments: 15 pages, 8 figures, 2 tables. arXiv admin note: text overlap with arXiv:2205.02712

Thermodynamics of the RN-AdS black hole with cloud of strings and quintessence in stationary and free-fall frame in rainbow gravity

Authors: Siyuan Hui, Benrong Mu, Yuzhou Tao, Jun Tao

Abstract: In this paper, we investigate the thermodynamic properties of the RN-AdS black hole with cloud of strings and quintessence in rainbow gravity with the stationary (ST) orthonormal frame and the free-fall (FF) orthonormal frame. After the SF and the FF rainbow metric is obtained, we get the Hawking temperature and the entropy, and their physical meanings are discussed. We find that, for the ST rainb… ▽ More In this paper, we investigate the thermodynamic properties of the RN-AdS black hole with cloud of strings and quintessence in rainbow gravity with the stationary (ST) orthonormal frame and the free-fall (FF) orthonormal frame. After the SF and the FF rainbow metric is obtained, we get the Hawking temperature and the entropy, and their physical meanings are discussed. We find that, for the ST rainbow RN-AdS black hole with cloud of strings and quintessence, the effect of rainbow gravity is to increase the Hawking temperature but decrease the entropy of the black hole. However, for the FF rainbow case, rainbow gravity turns out to decrease the Hawking temperature but increase the entropy of the black hole, which seems that the effects rainbow gravity has are quite model-dependent. △ Less

Submitted 5 May, 2022; originally announced May 2022.

Comments: 14 pages, 5 figures, 1 table

Machine Learning for Food Review and Recommendation

Authors: Tan Khang Le, Siu Cheung Hui

Abstract: Food reviews and recommendations have always been important for online food service websites. However, reviewing and recommending food is not simple as it is likely to be overwhelmed by disparate contexts and meanings. In this paper, we use different deep learning approaches to address the problems of sentiment analysis, automatic review tag generation, and retrieval of food reviews. We propose to… ▽ More Food reviews and recommendations have always been important for online food service websites. However, reviewing and recommending food is not simple as it is likely to be overwhelmed by disparate contexts and meanings. In this paper, we use different deep learning approaches to address the problems of sentiment analysis, automatic review tag generation, and retrieval of food reviews. We propose to develop a web-based food review system at Nanyang Technological University (NTU) named NTU Food Hunter, which incorporates different deep learning approaches that help users with food selection. First, we implement the BERT and LSTM deep learning models into the system for sentiment analysis of food reviews. Then, we develop a Part-of-Speech (POS) algorithm to automatically identify and extract adjective-noun pairs from the review content for review tag generation based on POS tagging and dependency parsing. Finally, we also train a RankNet model for the re-ranking of the retrieval results to improve the accuracy in our Solr-based food reviews search system. The experimental results show that our proposed deep learning approaches are promising for the applications of real-world problems. △ Less

Submitted 14 January, 2022; originally announced January 2022.

Comments: Accepted paper to International Student Conference on Artificial Intelligence (STCAI) 2021

Auxiliary Loss Reweighting for Image Inpainting

Authors: Siqi Hui, San** Zhou, Ye Deng, Wenli Huang, **jun Wang

Abstract: Image Inpainting is a task that aims to fill in missing regions of corrupted images with plausible contents. Recent inpainting methods have introduced perceptual and style losses as auxiliary losses to guide the learning of inpainting generators. Perceptual and style losses help improve the perceptual quality of inpainted results by supervising deep features of generated regions. However, two chal… ▽ More Image Inpainting is a task that aims to fill in missing regions of corrupted images with plausible contents. Recent inpainting methods have introduced perceptual and style losses as auxiliary losses to guide the learning of inpainting generators. Perceptual and style losses help improve the perceptual quality of inpainted results by supervising deep features of generated regions. However, two challenges have emerged with the usage of auxiliary losses: (i) the time-consuming grid search is required to decide weights for perceptual and style losses to properly perform, and (ii) loss terms with different auxiliary abilities are equally weighted by perceptual and style losses. To meet these two challenges, we propose a novel framework that independently weights auxiliary loss terms and adaptively adjusts their weights within a single training process, without a time-consuming grid search. Specifically, to release the auxiliary potential of perceptual and style losses, we propose two auxiliary losses, Tunable Perceptual Loss (TPL) and Tunable Style Loss (TSL) by using different tunable weights to consider the contributions of different loss terms. TPL and TSL are supersets of perceptual and style losses and release the auxiliary potential of standard perceptual and style losses. We further propose the Auxiliary Weights Adaptation (AWA) algorithm, which efficiently reweights TPL and TSL in a single training process. AWA is based on the principle that the best auxiliary weights would lead to the most improvement in inpainting performance. We conduct experiments on publically available datasets and find that our framework helps current SOTA methods achieve better results. △ Less

Submitted 22 April, 2022; v1 submitted 14 November, 2021; originally announced November 2021.

Dynamic Response and Stability Margin Improvement of Wireless Power Receiver Systems via Right-Half-Plane Zero Elimination

Authors: Kerui Li, Siew-Chong Tan, Ron Shu Yuen Hui

Abstract: The series-series compensation topology is widely adopted in many wireless power transfer applications. For such systems, their wireless power receiver part typically involves a DC-DC converter with front-stage full-bridge diode rectifier, to process the high-frequency transmitted AC power into a DC output voltage for the load. It is recently reported that the current source nature of the series-s… ▽ More The series-series compensation topology is widely adopted in many wireless power transfer applications. For such systems, their wireless power receiver part typically involves a DC-DC converter with front-stage full-bridge diode rectifier, to process the high-frequency transmitted AC power into a DC output voltage for the load. It is recently reported that the current source nature of the series-series compensation will introduce right-half-plane (RHP) zeros into the small-signal transfer functions of the DC-DC converter of the wireless power receiver, which will severely affect the stability and dynamic response of the system. To resolve this issue, in this paper, it is proposed to adopt a different rectifier configuration for the system such that the input current to the DC-DC converter becomes controllable to eliminate the presence of RHP zeros of the small-signal transfer functions of the system. This rectifier can be applied to different wireless power receivers using the buck, buck-boost, or boost converters. As compared with the original wireless power receivers, the modified ones feature minimum-phase characteristics and hence ease the design of compensator. Theoretical and experimental results are provided. The comparative experimental results verify the elimination of the RHP zero, improved dynamic responses of reference tracking and against load disturbances, and a larger stability margin. △ Less

Submitted 17 April, 2021; originally announced June 2021.

Comments: IEEE Transactions on Power Electronics, 2021

Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters

Authors: Aston Zhang, Yi Tay, Shuai Zhang, Alvin Chan, Anh Tuan Luu, Siu Cheung Hui, Jie Fu

Abstract: Recent works have demonstrated reasonable success of representation learning in hypercomplex space. Specifically, "fully-connected layers with Quaternions" (4D hypercomplex numbers), which replace real-valued matrix multiplications in fully-connected layers with Hamilton products of Quaternions, both enjoy parameter savings with only 1/4 learnable parameters and achieve comparable performance in v… ▽ More Recent works have demonstrated reasonable success of representation learning in hypercomplex space. Specifically, "fully-connected layers with Quaternions" (4D hypercomplex numbers), which replace real-valued matrix multiplications in fully-connected layers with Hamilton products of Quaternions, both enjoy parameter savings with only 1/4 learnable parameters and achieve comparable performance in various applications. However, one key caveat is that hypercomplex space only exists at very few predefined dimensions (4D, 8D, and 16D). This restricts the flexibility of models that leverage hypercomplex multiplications. To this end, we propose parameterizing hypercomplex multiplications, allowing models to learn multiplication rules from data regardless of whether such rules are predefined. As a result, our method not only subsumes the Hamilton product, but also learns to operate on any arbitrary nD hypercomplex space, providing more architectural flexibility using arbitrarily $1/n$ learnable parameters compared with the fully-connected layer counterpart. Experiments of applications to the LSTM and Transformer models on natural language inference, machine translation, text style transfer, and subject verb agreement demonstrate architectural flexibility and effectiveness of the proposed approach. △ Less

Submitted 17 February, 2021; originally announced February 2021.

Comments: Published as a conference paper at the 9th International Conference on Learning Representations (ICLR 2021)

On Effect of Right-Half-Plane Zero Present in Buck Converters with Input Current Source in Wireless Power Receiver Systems

Authors: Kerui Li, Siew-Chong Tan, Ron Shu Yuen Hui

Abstract: In wireless power receiver systems, the buck converter is widely used to step down the higher rectified voltage derived from the wireless receiver coil, to a lower output voltage for the immediate battery charging process. In this work, the presence and effect of the right-half-plane (RHP) zeros found in the small-signal inductor-current-to-duty-ratio and output-voltage-to-duty ratio transfer func… ▽ More In wireless power receiver systems, the buck converter is widely used to step down the higher rectified voltage derived from the wireless receiver coil, to a lower output voltage for the immediate battery charging process. In this work, the presence and effect of the right-half-plane (RHP) zeros found in the small-signal inductor-current-to-duty-ratio and output-voltage-to-duty ratio transfer functions of the buck converter in the wireless power receiver system on the control performance, are investigated. It is found and mathematically proved that the RHP zeros are introduced by the current source nature of the system attributed to the series-series compensation and finite DC-link capacitance. The RHP zero not only results in non-monotonic open-loop dynamic response but also complicates the design of feedback control and causes potential closed-loop instability. Theoretical and experimental results are provided to validate the presence of the RHP zeros and their effect on open-loop and closed-loop dynamic responses. △ Less

Submitted 25 November, 2020; originally announced November 2020.

Comments: 11 pages. IEEE Transactions on Power Electronics 2020 (Early access)

Journal ref: IEEE Trans. Power Electron., vol. 36, no. 6, pp. 6364-6374, June 2021

Value of Point-of-load Voltage Control for Enhanced Frequency Response in Future GB Power System

Authors: **rui Guo, Luis Badesa, Fei Teng, Balarko Chaudhuri, Shu Yuen Ron Hui, Goran Strbac

Abstract: The need for Enhanced Frequency Response (EFR) is expected to increase significantly in future low-carbon Great Britain (GB) power system. One way to provide EFR is to use power electronic compensators (PECs) for point-of-load voltage control (PVC) to exploit the voltage dependence of loads. This paper investigates the techno-economic feasibility of such technology in future GB power system by qua… ▽ More The need for Enhanced Frequency Response (EFR) is expected to increase significantly in future low-carbon Great Britain (GB) power system. One way to provide EFR is to use power electronic compensators (PECs) for point-of-load voltage control (PVC) to exploit the voltage dependence of loads. This paper investigates the techno-economic feasibility of such technology in future GB power system by quantifying the total EFR obtainable through deploying PVC in the urban domestic sector, the investment cost of the installment and the economic and environmental benefits of using PVC. The quantification is based on a stochastic domestic demand model and generic medium and low-voltage distribution networks for the urban areas of GB and a stochastic unit commitment (SUC) model with constraints for secure post-fault frequency evolution is used for the value assessment. Two future energy scenarios in the backdrop of 2030 with `smart' and `non-smart' control of electric vehicles and heat pumps, under different levels of penetration of battery energy storage system (BESS) are considered to assess the value of PEC, as well as the associated payback period. It is demonstrated that PVC could effectively complement BESS towards EFR provision in future GB power system. △ Less

Submitted 5 June, 2020; originally announced June 2020.

Comments: IEEE Transactions on Smart Grid

Journal ref: IEEE Transactions on Smart Grid, vol. 11, no. 6, pp. 4938-4948, Nov. 2020

On Beat Frequency Oscillation of Two-Stage Wireless Power Receivers

Authors: Kerui Li, Siew-Chong Tan, Ron Shu Yuen Hui

Abstract: Two-stage wireless power receivers, which typically include an AC-DC diode rectifier and a DC-DC regulator, are popular solutions in low-power wireless power transfer applications. However, the interaction between the rectifier and the regulator may introduce beat frequency oscillation on both the DC-link and output capacitors. In this paper, the cause of the beat frequency oscillation and its rel… ▽ More Two-stage wireless power receivers, which typically include an AC-DC diode rectifier and a DC-DC regulator, are popular solutions in low-power wireless power transfer applications. However, the interaction between the rectifier and the regulator may introduce beat frequency oscillation on both the DC-link and output capacitors. In this paper, the cause of the beat frequency oscillation and its related issues are investigated with the corresponding design solution on alleviating the oscillation discussed. Theoretical and experimental results verifying the presence of beat frequency oscillation in the two-stage wireless receiver system are provided. Our study shows that the beat frequency oscillation can be significantly alleviated if appropriate design solutions are applied. △ Less

Submitted 5 October, 2020; v1 submitted 28 April, 2020; originally announced April 2020.

Journal ref: in IEEE Transactions on Power Electronics, vol. 35, no. 12, pp. 12741-12751, Dec. 2020

Highly-Efficient Single-Switch-Regulated Resonant Wireless Power Receiver with Hybrid Modulation

Authors: Kerui Li, Albert Ting Leung Lee, Siew-Chong Tan, Ron Shu Yuen Hui

Abstract: In this paper, a highly-efficient single-switch-regulated resonant wireless power receiver with hybrid modulation is proposed. To achieve both high efficiency and good output voltage regulation, phase shift and pulse width hybrid modulation are simultaneously applied. The soft switching operation in this topology is achieved by the cycle-by-cycle phase shift adjustment between the input current an… ▽ More In this paper, a highly-efficient single-switch-regulated resonant wireless power receiver with hybrid modulation is proposed. To achieve both high efficiency and good output voltage regulation, phase shift and pulse width hybrid modulation are simultaneously applied. The soft switching operation in this topology is achieved by the cycle-by-cycle phase shift adjustment between the input current and the gate drive signal and also attributed to the reactive components such as the series-compensated secondary coil and the parasitic capacitor of the active switch . The soft switching operation also leads to high efficiency and low EMI. By adjusting the duty ratio of the switch, tight regulation of the output voltage can be attained. The steady-state and dynamic models of the resonant receiver with hybrid modulation are analytically derived in order to properly design the feedback controller. An experimental setup of a two-coil wireless power transfer system, including the hardware prototype of the proposed receiver, is constructed for experimental verification. The experimental results show the effectiveness of the soft-switching operation in the receiver with high efficiency while maintaining good regulation of the output voltage, regardless of line and load variations. △ Less

Submitted 5 January, 2021; v1 submitted 9 April, 2020; originally announced April 2020.

Comments: in IEEE Journal of Emerging and Selected Topics in Power Electronics. 2020

Single-Switch-Regulated Resonant WPT Receiver

Authors: Kerui Li, Siew Chong Tan, Ron Shu Yuen Hui

Abstract: A single-switch-regulated wireless power transfer (WPT) receiver is presented in this letter. Aiming at low-cost applications, the system involves only a single-switch class-E resonant rectifier, a frequency synchronization circuit, and a microcontroller. The number of power semiconductor devices required in this circuit is minimal. Only one active switch is used and no diode is required. As a sin… ▽ More A single-switch-regulated wireless power transfer (WPT) receiver is presented in this letter. Aiming at low-cost applications, the system involves only a single-switch class-E resonant rectifier, a frequency synchronization circuit, and a microcontroller. The number of power semiconductor devices required in this circuit is minimal. Only one active switch is used and no diode is required. As a single-switch solution, this simplifies circuit implementation, improves reliability, and lowers hardware cost. The single-switch resonant rectifier provides a relatively constant quasi-sinusoidal voltage waveform to pick up the wireless power from the receiver coil. Due to the resonant nature of the rectifier, ZVS turn on and turn off are achieved. The steady-state analysis and discussions on the component sizing and the control design are provided. A prototype is built and experimental works are performed to verify the features. △ Less

Submitted 18 December, 2019; v1 submitted 12 December, 2019; originally announced December 2019.

Journal ref: in IEEE Transactions on Power Electronics, vol. 34, no. 11, pp. 10386-10391, Nov. 2019

Single-Stage Regulated Resonant WPT Receiver with Low Input Harmonic Distortion

Authors: Kerui Li, Siew Chong Tan, Ron Shu Yuen Hui

Abstract: Resonant rectifier topologies would be a promising candidate for achieving simple, compact, and reliable single-stage wireless power transfer (WPT) receiver if not for the lack of good DC regulation capability. This paper investigates the problems that prevent the feasibility of single-stage DC regulation in resonant rectifier topologies. A possible solution is the proposed differential resonant r… ▽ More Resonant rectifier topologies would be a promising candidate for achieving simple, compact, and reliable single-stage wireless power transfer (WPT) receiver if not for the lack of good DC regulation capability. This paper investigates the problems that prevent the feasibility of single-stage DC regulation in resonant rectifier topologies. A possible solution is the proposed differential resonant rectifier topology, of which the rectifier is designed to have a relatively constant AC voltage, and that phase shift control is used to achieve relatively good output regulation. Design considerations on the reactive component sizing, magnetic component design, frequency and phase synchronization, small signal modelling, and closed-loop feedback control design, are discussed. Experimental results verified that the proposed WPT receiver system can achieve single-stage AC rectification and DC regulation while attaining the key features of low harmonic distortion in its AC output voltage, continuous DC current, and zero-voltage-switching (ZVS) operation over a wide operating range. △ Less

Submitted 6 January, 2020; v1 submitted 12 December, 2019; originally announced December 2019.

Journal ref: in IEEE Transactions on Power Electronics, vol. 35, no. 7, pp. 6820-6829, July 2020

*-Conformal η-Ricci Soliton on Sasakian manifold

Authors: Soumendu Roy, Santu Dey, Arindam Bhattacharyya, Shyamal Kumar Hui

Abstract: In this paper we study *-Conformal η-Ricci soliton on Sasakian manifolds. Here, we discuss some curvature properties on Sasakian manifold admitting *-Conformal η-Ricci soliton. We obtain some significant results on *-Conformal η-Ricci soliton in Sasakian manifolds satisfying R(ξ,X).S = 0, S(ξ,X).R = 0, {\overline}P(ξ,X).S = 0, where {\overline}P is Pseudo-projective curvature tensor.The conditions… ▽ More In this paper we study *-Conformal η-Ricci soliton on Sasakian manifolds. Here, we discuss some curvature properties on Sasakian manifold admitting *-Conformal η-Ricci soliton. We obtain some significant results on *-Conformal η-Ricci soliton in Sasakian manifolds satisfying R(ξ,X).S = 0, S(ξ,X).R = 0, {\overline}P(ξ,X).S = 0, where {\overline}P is Pseudo-projective curvature tensor.The conditions for *-Conformal η-Ricci soliton on Φ-conharmonically flat and Φ-projectively flat Sasakian manifolds have been obtained in this article. Lastly we give an example of 5-dimensional Sasakian manifolds satisfying *-Conformal η-Ricci soliton. △ Less

Submitted 3 September, 2019; originally announced September 2019.

Comments: 15 pages

MSC Class: 53A30; 53C15; 53C25; 53D15

Journal ref: 2021

MRI Reconstruction Using Deep Bayesian Estimation

Authors: GuanXiong Luo, Na Zhao, Wenhao Jiang, Edward S. Hui, Peng Cao

Abstract: Purpose: To develop a deep learning-based Bayesian inference for MRI reconstruction. Methods: We modeled the MRI reconstruction problem with Bayes's theorem, following the recently proposed PixelCNN++ method. The image reconstruction from incomplete k-space measurement was obtained by maximizing the posterior possibility. A generative network was utilized as the image prior, which was computationa… ▽ More Purpose: To develop a deep learning-based Bayesian inference for MRI reconstruction. Methods: We modeled the MRI reconstruction problem with Bayes's theorem, following the recently proposed PixelCNN++ method. The image reconstruction from incomplete k-space measurement was obtained by maximizing the posterior possibility. A generative network was utilized as the image prior, which was computationally tractable, and the k-space data fidelity was enforced by using an equality constraint. The stochastic backpropagation was utilized to calculate the descent gradient in the process of maximum a posterior, and a projected subgradient method was used to impose the equality constraint. In contrast to the other deep learning reconstruction methods, the proposed one used the likelihood of prior as the training loss and the objective function in reconstruction to improve the image quality. Results: The proposed method showed an improved performance in preserving image details and reducing aliasing artifacts, compared with GRAPPA, $\ell_1$-ESPRiT, and MODL, a state-of-the-art deep learning reconstruction method. The proposed method generally achieved more than 5 dB peak signal-to-noise ratio improvement for compressed sensing and parallel imaging reconstructions compared with the other methods. Conclusion: The Bayesian inference significantly improved the reconstruction performance, compared with the conventional $\ell_1$-sparsity prior in compressed sensing reconstruction tasks. More importantly, the proposed reconstruction framework can be generalized for most MRI reconstruction scenarios. △ Less

Submitted 17 February, 2022; v1 submitted 3 September, 2019; originally announced September 2019.

Collecting and Analyzing Multidimensional Data with Local Differential Privacy

Authors: Ning Wang, Xiaokui Xiao, Yin Yang, Jun Zhao, Siu Cheung Hui, Hye** Shin, Junbum Shin, Ge Yu

Abstract: Local differential privacy (LDP) is a recently proposed privacy standard for collecting and analyzing data, which has been used, e.g., in the Chrome browser, iOS and macOS. In LDP, each user perturbs her information locally, and only sends the randomized version to an aggregator who performs analyses, which protects both the users and the aggregator against private information leaks. Although LDP… ▽ More Local differential privacy (LDP) is a recently proposed privacy standard for collecting and analyzing data, which has been used, e.g., in the Chrome browser, iOS and macOS. In LDP, each user perturbs her information locally, and only sends the randomized version to an aggregator who performs analyses, which protects both the users and the aggregator against private information leaks. Although LDP has attracted much research attention in recent years, the majority of existing work focuses on applying LDP to complex data and/or analysis tasks. In this paper, we point out that the fundamental problem of collecting multidimensional data under LDP has not been addressed sufficiently, and there remains much room for improvement even for basic tasks such as computing the mean value over a single numeric attribute under LDP. Motivated by this, we first propose novel LDP mechanisms for collecting a numeric attribute, whose accuracy is at least no worse (and usually better) than existing solutions in terms of worst-case noise variance. Then, we extend these mechanisms to multidimensional data that can contain both numeric and categorical attributes, where our mechanisms always outperform existing solutions regarding worst-case noise variance. As a case study, we apply our solutions to build an LDP-compliant stochastic gradient descent algorithm (SGD), which powers many important machine learning tasks. Experiments using real datasets confirm the effectiveness of our methods, and their advantages over existing solutions. △ Less

Submitted 28 June, 2019; originally announced July 2019.

Comments: 12-Page Full Paper in Proceedings of the 2019 IEEE International Conference on Data Engineering (ICDE). arXiv admin note: text overlap with arXiv:1606.05053

MSC Class: Local differential privacy; multidimensional data; stochastic gradient descent

Lightweight and Efficient Neural Natural Language Processing with Quaternion Networks

Authors: Yi Tay, Aston Zhang, Luu Anh Tuan, **feng Rao, Shuai Zhang, Shuohang Wang, Jie Fu, Siu Cheung Hui

Abstract: Many state-of-the-art neural models for NLP are heavily parameterized and thus memory inefficient. This paper proposes a series of lightweight and memory efficient neural architectures for a potpourri of natural language processing (NLP) tasks. To this end, our models exploit computation using Quaternion algebra and hypercomplex spaces, enabling not only expressive inter-component interactions but… ▽ More Many state-of-the-art neural models for NLP are heavily parameterized and thus memory inefficient. This paper proposes a series of lightweight and memory efficient neural architectures for a potpourri of natural language processing (NLP) tasks. To this end, our models exploit computation using Quaternion algebra and hypercomplex spaces, enabling not only expressive inter-component interactions but also significantly ($75\%$) reduced parameter size due to lesser degrees of freedom in the Hamilton product. We propose Quaternion variants of models, giving rise to new architectures such as the Quaternion attention Model and Quaternion Transformer. Extensive experiments on a battery of NLP tasks demonstrates the utility of proposed Quaternion-inspired models, enabling up to $75\%$ reduction in parameter size without significant loss in performance. △ Less

Submitted 11 June, 2019; originally announced June 2019.

Comments: ACL 2019

Simple and Effective Curriculum Pointer-Generator Networks for Reading Comprehension over Long Narratives

Authors: Yi Tay, Shuohang Wang, Luu Anh Tuan, Jie Fu, Minh C. Phan, Xingdi Yuan, **feng Rao, Siu Cheung Hui, Aston Zhang

Abstract: This paper tackles the problem of reading comprehension over long narratives where documents easily span over thousands of tokens. We propose a curriculum learning (CL) based Pointer-Generator framework for reading/sampling over large documents, enabling diverse training of the neural model based on the notion of alternating contextual difficulty. This can be interpreted as a form of domain random… ▽ More This paper tackles the problem of reading comprehension over long narratives where documents easily span over thousands of tokens. We propose a curriculum learning (CL) based Pointer-Generator framework for reading/sampling over large documents, enabling diverse training of the neural model based on the notion of alternating contextual difficulty. This can be interpreted as a form of domain randomization and/or generative pretraining during training. To this end, the usage of the Pointer-Generator softens the requirement of having the answer within the context, enabling us to construct diverse training samples for learning. Additionally, we propose a new Introspective Alignment Layer (IAL), which reasons over decomposed alignments using block-based self-attention. We evaluate our proposed method on the NarrativeQA reading comprehension benchmark, achieving state-of-the-art performance, improving existing baselines by $51\%$ relative improvement on BLEU-4 and $17\%$ relative improvement on Rouge-L. Extensive ablations confirm the effectiveness of our proposed IAL and CL components. △ Less

Submitted 26 May, 2019; originally announced May 2019.

Comments: Accepted to ACL 2019

Invariant Submanifolds of Sasakian Generalized-Sasakian-Space-Forms

Authors: D. G. Prakasha, P. Veeresha, Inan Unal, Shyamal Kumar Hui

Abstract: The object of this paper is to study the invariant submanifolds of Sasakian generalized-Sasakian-space-form. Here, we obtain some equivalent conditions for an invariant submanifold of a Sasakian generalized-Sasakian-space-forms to be totally geodesic. The object of this paper is to study the invariant submanifolds of Sasakian generalized-Sasakian-space-form. Here, we obtain some equivalent conditions for an invariant submanifold of a Sasakian generalized-Sasakian-space-forms to be totally geodesic. △ Less

Submitted 9 December, 2022; v1 submitted 30 January, 2019; originally announced January 2019.

Comments: 11 pages

MSC Class: 53C15; 53C40

Certain results on (LCS)n-manifolds

Authors: Vishnuvardhana. S. V., Venkatesha, S. K. Hui

Abstract: The purpose of the present paper is to study semi-generalized recurrent, semi-generalized Ricci recurrent and conformal Ricci soliton on (LCS)n-manifold. The purpose of the present paper is to study semi-generalized recurrent, semi-generalized Ricci recurrent and conformal Ricci soliton on (LCS)n-manifold. △ Less

Submitted 21 January, 2019; originally announced January 2019.

MSC Class: 53C15; 53C25; 53C50

Recurrently Controlled Recurrent Networks

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui

Abstract: Recurrent neural networks (RNNs) such as long short-term memory and gated recurrent units are pivotal building blocks across a broad spectrum of sequence modeling problems. This paper proposes a recurrently controlled recurrent network (RCRN) for expressive and powerful sequence encoding. More concretely, the key idea behind our approach is to learn the recurrent gating functions using recurrent n… ▽ More Recurrent neural networks (RNNs) such as long short-term memory and gated recurrent units are pivotal building blocks across a broad spectrum of sequence modeling problems. This paper proposes a recurrently controlled recurrent network (RCRN) for expressive and powerful sequence encoding. More concretely, the key idea behind our approach is to learn the recurrent gating functions using recurrent networks. Our architecture is split into two components - a controller cell and a listener cell whereby the recurrent controller actively influences the compositionality of the listener cell. We conduct extensive experiments on a myriad of tasks in the NLP domain such as sentiment analysis (SST, IMDb, Amazon reviews, etc.), question classification (TREC), entailment classification (SNLI, SciTail), answer selection (WikiQA, TrecQA) and reading comprehension (NarrativeQA). Across all 26 datasets, our results demonstrate that RCRN not only consistently outperforms BiLSTMs but also stacked BiLSTMs, suggesting that our controller architecture might be a suitable replacement for the widely adopted stacked architecture. △ Less

Submitted 24 November, 2018; originally announced November 2018.

Comments: NIPS 2018

Densely Connected Attention Propagation for Reading Comprehension

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui, Jian Su

Abstract: We propose DecaProp (Densely Connected Attention Propagation), a new densely connected neural architecture for reading comprehension (RC). There are two distinct characteristics of our model. Firstly, our model densely connects all pairwise layers of the network, modeling relationships between passage and query across all hierarchical levels. Secondly, the dense connectors in our network are learn… ▽ More We propose DecaProp (Densely Connected Attention Propagation), a new densely connected neural architecture for reading comprehension (RC). There are two distinct characteristics of our model. Firstly, our model densely connects all pairwise layers of the network, modeling relationships between passage and query across all hierarchical levels. Secondly, the dense connectors in our network are learned via attention instead of standard residual skip-connectors. To this end, we propose novel Bidirectional Attention Connectors (BAC) for efficiently forging connections throughout the network. We conduct extensive experiments on four challenging RC benchmarks. Our proposed approach achieves state-of-the-art results on all four, outperforming existing baselines by up to $2.6\%-14.2\%$ in absolute F1 score. △ Less

Submitted 2 April, 2019; v1 submitted 10 November, 2018; originally announced November 2018.

Comments: NIPS 2018

Co-Stack Residual Affinity Networks with Multi-level Attention Refinement for Matching Text Sequences

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui

Abstract: Learning a matching function between two text sequences is a long standing problem in NLP research. This task enables many potential applications such as question answering and paraphrase identification. This paper proposes Co-Stack Residual Affinity Networks (CSRAN), a new and universal neural architecture for this problem. CSRAN is a deep architecture, involving stacked (multi-layered) recurrent… ▽ More Learning a matching function between two text sequences is a long standing problem in NLP research. This task enables many potential applications such as question answering and paraphrase identification. This paper proposes Co-Stack Residual Affinity Networks (CSRAN), a new and universal neural architecture for this problem. CSRAN is a deep architecture, involving stacked (multi-layered) recurrent encoders. Stacked/Deep architectures are traditionally difficult to train, due to the inherent weaknesses such as difficulty with feature propagation and vanishing gradients. CSRAN incorporates two novel components to take advantage of the stacked architecture. Firstly, it introduces a new bidirectional alignment mechanism that learns affinity weights by fusing sequence pairs across stacked hierarchies. Secondly, it leverages a multi-level attention refinement component between stacked recurrent layers. The key intuition is that, by leveraging information across all network hierarchies, we can not only improve gradient flow but also improve overall performance. We conduct extensive experiments on six well-studied text sequence matching datasets, achieving state-of-the-art performance on all. △ Less

Submitted 6 October, 2018; originally announced October 2018.

Comments: EMNLP 2018

Another class of warped product skew CR-submanifolds of Kenmotsu manifolds

Authors: Shyamal Kumar Hui, Tanumoy Pal, Joydeb Roy

Abstract: Recently, Naghi et al. \cite{NAGHI} studied warped product skew CR-submanifold of the form $M_1\times_fM_\bot$ of order $1$ of a Kenmotsu manifold $\bar{M}$ such that $M_1=M_T\times M_θ$, where $M_T$, $M_\bot$ and $M_θ$ are invariant, anti-invariant and proper slant submanifolds of $\bar{M}$. The present paper deals with the study of warped product submanifolds by interchanging the two factors… ▽ More Recently, Naghi et al. \cite{NAGHI} studied warped product skew CR-submanifold of the form $M_1\times_fM_\bot$ of order $1$ of a Kenmotsu manifold $\bar{M}$ such that $M_1=M_T\times M_θ$, where $M_T$, $M_\bot$ and $M_θ$ are invariant, anti-invariant and proper slant submanifolds of $\bar{M}$. The present paper deals with the study of warped product submanifolds by interchanging the two factors $M_T$ and $M_\bot$, i.e, the warped products of the form $M_2\times_fM_T$ such that $M_2=M_\bot\times M_θ$. The existence of such warped product is ensured by an example and then we characterize such warped product submanifold. A lower bounds of the square norm of second fundamental form is derived with sharp relation, whose equality case is also considered. △ Less

Submitted 26 June, 2018; originally announced June 2018.

Comments: 20 pages

Self-Attentive Neural Collaborative Filtering

Authors: Yi Tay, Shuai Zhang, Luu Anh Tuan, Siu Cheung Hui

Abstract: This paper has been withdrawn as we discovered a bug in our tensorflow implementation that involved accidental mixing of vectors across batches. This lead to different inference results given different batch sizes which is completely strange. The performance scores still remain the same but we concluded that it was not the self-attention that contributed to the performance. We are withdrawing the… ▽ More This paper has been withdrawn as we discovered a bug in our tensorflow implementation that involved accidental mixing of vectors across batches. This lead to different inference results given different batch sizes which is completely strange. The performance scores still remain the same but we concluded that it was not the self-attention that contributed to the performance. We are withdrawing the paper because this renders the main claim of the paper false. Thanks to Guan Xinyu from NUS for discovering this issue in our previously open source code. △ Less

Submitted 19 July, 2018; v1 submitted 17 June, 2018; originally announced June 2018.

Comments: We discovered a bug in our tensorflow implementation that involved accidental mixing of vectors across batches, rendering the main claim of the paper incorrect. We are withdrawing this paper until we find out why

Multi-Cast Attention Networks for Retrieval-based Question Answering and Response Prediction

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui

Abstract: Attention is typically used to select informative sub-phrases that are used for prediction. This paper investigates the novel use of attention as a form of feature augmentation, i.e, casted attention. We propose Multi-Cast Attention Networks (MCAN), a new attention mechanism and general model architecture for a potpourri of ranking tasks in the conversational modeling and question answering domain… ▽ More Attention is typically used to select informative sub-phrases that are used for prediction. This paper investigates the novel use of attention as a form of feature augmentation, i.e, casted attention. We propose Multi-Cast Attention Networks (MCAN), a new attention mechanism and general model architecture for a potpourri of ranking tasks in the conversational modeling and question answering domains. Our approach performs a series of soft attention operations, each time casting a scalar feature upon the inner word embeddings. The key idea is to provide a real-valued hint (feature) to a subsequent encoder layer and is targeted at improving the representation learning process. There are several advantages to this design, e.g., it allows an arbitrary number of attention mechanisms to be casted, allowing for multiple attention types (e.g., co-attention, intra-attention) and attention variants (e.g., alignment-pooling, max-pooling, mean-pooling) to be executed simultaneously. This not only eliminates the costly need to tune the nature of the co-attention layer, but also provides greater extents of explainability to practitioners. Via extensive experiments on four well-known benchmark datasets, we show that MCAN achieves state-of-the-art performance. On the Ubuntu Dialogue Corpus, MCAN outperforms existing state-of-the-art models by $9\%$. MCAN also achieves the best performing score to date on the well-studied TrecQA dataset. △ Less

Submitted 3 June, 2018; originally announced June 2018.

Comments: Accepted to KDD 2018 (Paper titled only "Multi-Cast Attention Networks" in KDD version)

CoupleNet: Paying Attention to Couples with Coupled Attention for Relationship Recommendation

Authors: Yi Tay, Anh Tuan Luu, Siu Cheung Hui

Abstract: Dating and romantic relationships not only play a huge role in our personal lives but also collectively influence and shape society. Today, many romantic partnerships originate from the Internet, signifying the importance of technology and the web in modern dating. In this paper, we present a text-based computational approach for estimating the relationship compatibility of two users on social med… ▽ More Dating and romantic relationships not only play a huge role in our personal lives but also collectively influence and shape society. Today, many romantic partnerships originate from the Internet, signifying the importance of technology and the web in modern dating. In this paper, we present a text-based computational approach for estimating the relationship compatibility of two users on social media. Unlike many previous works that propose reciprocal recommender systems for online dating websites, we devise a distant supervision heuristic to obtain real world couples from social platforms such as Twitter. Our approach, the CoupleNet is an end-to-end deep learning based estimator that analyzes the social profiles of two users and subsequently performs a similarity match between the users. Intuitively, our approach performs both user profiling and match-making within a unified end-to-end framework. CoupleNet utilizes hierarchical recurrent neural models for learning representations of user profiles and subsequently coupled attention mechanisms to fuse information aggregated from two users. To the best of our knowledge, our approach is the first data-driven deep learning approach for our novel relationship recommendation problem. We benchmark our CoupleNet against several machine learning and deep learning baselines. Experimental results show that our approach outperforms all approaches significantly in terms of precision. Qualitative analysis shows that our model is capable of also producing explainable results to users. △ Less

Submitted 29 May, 2018; originally announced May 2018.

Comments: Accepted at ICWSM 2018

Reasoning with Sarcasm by Reading In-between

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui, Jian Su

Abstract: Sarcasm is a sophisticated speech act which commonly manifests on social communities such as Twitter and Reddit. The prevalence of sarcasm on the social web is highly disruptive to opinion mining systems due to not only its tendency of polarity flip** but also usage of figurative language. Sarcasm commonly manifests with a contrastive theme either between positive-negative sentiments or between… ▽ More Sarcasm is a sophisticated speech act which commonly manifests on social communities such as Twitter and Reddit. The prevalence of sarcasm on the social web is highly disruptive to opinion mining systems due to not only its tendency of polarity flip** but also usage of figurative language. Sarcasm commonly manifests with a contrastive theme either between positive-negative sentiments or between literal-figurative scenarios. In this paper, we revisit the notion of modeling contrast in order to reason with sarcasm. More specifically, we propose an attention-based neural model that looks in-between instead of across, enabling it to explicitly model contrast and incongruity. We conduct extensive experiments on six benchmark datasets from Twitter, Reddit and the Internet Argument Corpus. Our proposed model not only achieves state-of-the-art performance on all datasets but also enjoys improved interpretability. △ Less

Submitted 8 May, 2018; originally announced May 2018.

Comments: Accepted to ACL2018

Warped product pointwise bi-slant submanifolds of Kenmotsu manifolds

Authors: Shyamal Kumar Hui, Joydeb Roy, Tanumoy Pal

Abstract: The present paper deals with the study of warped product pointwise bi-slant submanifolds of Kenmotsu manifolds with an example. The characterization for such submanifold is also discussed. An inequality of such submanifold is obtained and its equality case is also considered. The present paper deals with the study of warped product pointwise bi-slant submanifolds of Kenmotsu manifolds with an example. The characterization for such submanifold is also discussed. An inequality of such submanifold is obtained and its equality case is also considered. △ Less

Submitted 28 March, 2018; originally announced March 2018.

Comments: 20 pages

MSC Class: 53C15; 53C42

Multi-range Reasoning for Machine Comprehension

Authors: Yi Tay, Luu Anh Tuan, Siu Cheung Hui

Abstract: We propose MRU (Multi-Range Reasoning Units), a new fast compositional encoder for machine comprehension (MC). Our proposed MRU encoders are characterized by multi-ranged gating, executing a series of parameterized contract-and-expand layers for learning gating vectors that benefit from long and short-term dependencies. The aims of our approach are as follows: (1) learning representations that are… ▽ More We propose MRU (Multi-Range Reasoning Units), a new fast compositional encoder for machine comprehension (MC). Our proposed MRU encoders are characterized by multi-ranged gating, executing a series of parameterized contract-and-expand layers for learning gating vectors that benefit from long and short-term dependencies. The aims of our approach are as follows: (1) learning representations that are concurrently aware of long and short-term context, (2) modeling relationships between intra-document blocks and (3) fast and efficient sequence encoding. We show that our proposed encoder demonstrates promising results both as a standalone encoder and as well as a complementary building block. We conduct extensive experiments on three challenging MC datasets, namely RACE, SearchQA and NarrativeQA, achieving highly competitive performance on all. On the RACE benchmark, our model outperforms DFN (Dynamic Fusion Networks) by 1.5%-6% without using any recurrent or convolution layers. Similarly, we achieve competitive performance relative to AMANDA on the SearchQA benchmark and BiDAF on the NarrativeQA benchmark without using any LSTM/GRU layers. Finally, incorporating MRU encoders with standard BiLSTM architectures further improves performance, achieving state-of-the-art results. △ Less

Submitted 24 March, 2018; originally announced March 2018.

A note on submanifolds of $\bar{M}^{2n+1}(f_1,f_2,f_3)$ with respect to certain connections

Authors: Pradip Mandal, Shyamal Kumar Hui

Abstract: The present paper deals with some results of almsot semi-invariant submanifolds of generalized Sasakian-space-forms in \cite{ALEGRE3} with respect to semisymmetric metric connection, semisymmetric non-metric connection, Schouten-van Kampen connection and Tanaka-Webster connection. The present paper deals with some results of almsot semi-invariant submanifolds of generalized Sasakian-space-forms in \cite{ALEGRE3} with respect to semisymmetric metric connection, semisymmetric non-metric connection, Schouten-van Kampen connection and Tanaka-Webster connection. △ Less

Submitted 15 March, 2018; originally announced March 2018.

Comments: 13 pages. arXiv admin note: text overlap with arXiv:1803.02610