arXiv subjects

Minh Nguyen

Publications and source records attributed to Minh Nguyen.

At least 91 records · Page 5Linked to original sources

arXiv2021

Linked visualisations via Galois dependencies

We present new language-based dynamic analysis techniques for linking visualisations and other structured outputs to data in a fine-grained way, allowing a user to interactively explore how data attributes map to visual or other output elements by selecting (focusing on) substructures of interest. This can help both programmers and end-users understand how data sources and complex outputs are related, which can be a challenge even for someone with expert knowledge of the problem domain. Our approach builds on bidirectional program slicing techiques based on Galois connections, which provide desirable round-tripping properties. Unlike the prior work in program slicing, our approach allows selections to be negated. In a setting with negation, the bidirectional analysis has a De Morgan dual, which can be used to link different outputs generated from the same input. This offers a principled language-based foundation for a popular interactive visualisation feature called brushing and linking where selections in one chart automatically select corresponding elements in another related chart. Although such view coordination features are valuable comprehension aids, they tend be to hard-coded into specific applications or libraries, or require programmer effort.

cs.PL↗

arXiv2021

Coupling spin defects in hexagonal boron nitride to monolithic bullseye cavities

Color centers in hexagonal boron nitride (hBN) are becoming an increasingly important building block for quantum photonic applications. Herein, we demonstrate the efficient coupling of recently discovered spin defects in hBN to purposely designed bullseye cavities. We show that the all monolithic hBN cavity system exhibits an order of magnitude enhancement in the emission of the coupled boron vacancy spin defects. In addition, by comparative finite difference time domain modelling, we shed light on the emission dipole orientation, which has not been experimentally demonstrated at this point. Beyond that, the coupled spin system exhibits an enhanced contrast in optically detected magnetic resonance readout and improved signal to noise ratio. Thus, our experimental results supported by simulations, constitute a first step towards integration of hBN spin defects with photonic resonators for a scalable spin photon interface.

physics.optics↗

arXiv2021

Domain-shift Conditioning using Adaptable Filtering via Hierarchical Embeddings for Robust Chinese Spell Check

Spell check is a useful application which processes noisy human-generated text. Spell check for Chinese poses unresolved problems due to the large number of characters, the sparse distribution of errors, and the dearth of resources with sufficient coverage of heterogeneous and shifting error domains. For Chinese spell check, filtering using confusion sets narrows the search space and makes finding corrections easier. However, most, if not all, confusion sets used to date are fixed and thus do not include new, shifting error domains. We propose a scalable adaptable filter that exploits hierarchical character embeddings to (1) obviate the need to handcraft confusion sets, and (2) resolve sparsity problems related to infrequent errors. Our approach compares favorably with competitive baselines and obtains SOTA results on the 2014 and 2015 Chinese Spelling Check Bake-off datasets.

cs.CL↗

arXiv2020

Gunrock 2.0: A User Adaptive Social Conversational System

Gunrock 2.0 is built on top of Gunrock with an emphasis on user adaptation. Gunrock 2.0 combines various neural natural language understanding modules, including named entity detection, linking, and dialog act prediction, to improve user understanding. Its dialog management is a hierarchical model that handles various topics, such as movies, music, and sports. The system-level dialog manager can handle question detection, acknowledgment, error handling, and additional functions, making downstream modules much easier to design and implement. The dialog manager also adapts its topic selection to accommodate different users' profile information, such as inferred gender and personality. The generation model is a mix of templates and neural generation models. Gunrock 2.0 is able to achieve an average rating of 3.73 at its latest build from May 29th to June 4th.

cs.CL↗

arXiv2020

Atomically Thin Boron Nitride as an Ideal Spacer for Metal-Enhanced Fluorescence

The metal-enhanced fluorescence (MEF) considerably enhances the luminescence for various applications, but its performance largely depends on the dielectric spacer between the fluorophore and plasmonic system. It is still challenging to produce a defect-free spacer having an optimized thickness with a subnanometer accuracy that enables reusability without affecting the enhancement. In this study, we demonstrate the use of atomically thin hexagonal boron nitride (BN) as an ideal MEF spacer owing to its multifold advantages over the traditional dielectric thin films. With rhodamine 6G as a representative fluorophore, it largely improves the enhancement factor (up to ~95+-5), sensitivity (10^-8 M), reproducibility, and reusability (~90% of the plasmonic activity is retained after 30 cycles of heating at 350 °C in air) of MEF. This can be attributed to its two-dimensional structure, thickness control at the atomic level, defect-free quality, high affinities to aromatic fluorophores, good thermal stability, and excellent impermeability. The atomically thin BN spacers could increase the use of MEF in different fields and industries.

physics.app-ph↗

arXiv2020

Image-based Vehicle Re-identification Model with Adaptive Attention Modules and Metadata Re-ranking

Vehicle Re-identification is a challenging task due to intra-class variability and inter-class similarity across non-overlapping cameras. To tackle these problems, recently proposed methods require additional annotation to extract more features for false positive image exclusion. In this paper, we propose a model powered by adaptive attention modules that requires fewer label annotations but still out-performs the previous models. We also include a re-ranking method that takes account of the importance of metadata feature embeddings in our paper. The proposed method is evaluated on CVPR AI City Challenge 2020 dataset and achieves mAP of 37.25% in Track 2.

cs.CV↗

arXiv2020

Hierarchical Character Embeddings: Learning Phonological and Semantic Representations in Languages of Logographic Origin using Recursive Neural Networks

Logographs (Chinese characters) have recursive structures (i.e. hierarchies of sub-units in logographs) that contain phonological and semantic information, as developmental psychology literature suggests that native speakers leverage on the structures to learn how to read. Exploiting these structures could potentially lead to better embeddings that can benefit many downstream tasks. We propose building hierarchical logograph (character) embeddings from logograph recursive structures using treeLSTM, a recursive neural network. Using recursive neural network imposes a prior on the mapping from logographs to embeddings since the network must read in the sub-units in logographs according to the order specified by the recursive structures. Based on human behavior in language learning and reading, we hypothesize that modeling logographs' structures using recursive neural network should be beneficial. To verify this claim, we consider two tasks (1) predicting logographs' Cantonese pronunciation from logographic structures and (2) language modeling. Empirical results show that the proposed hierarchical embeddings outperform baseline approaches. Diagnostic analysis suggests that hierarchical embeddings constructed using treeLSTM is less sensitive to distractors, thus is more robust, especially on complex logographs.

cs.CL↗

arXiv2019

Integrated on chip platform with quantum emitters in layered materials

Integrated quantum photonic circuitry is an emerging topic that requires efficient coupling of quantum light sources to waveguides and optical resonators. So far, great effort has been devoted to engineering on-chip systems from three-dimensional crystals such as diamond or gallium arsenide. In this study, we demonstrate room temperature coupling of quantum emitters embedded within a layered hexagonal boron nitride to an on-chip aluminium nitride waveguide. We achieved 1.2% light coupling efficiency of the device and realise transmission of single photons through the waveguide. Our results serve as a foundation for the integration of layered materials with on-chip components and for the realisation of integrated quantum photonic circuitry.

physics.app-ph↗

arXiv2019

Revealing multiple classes of stable quantum emitters in hexagonal boron nitride with correlated cathodoluminescence, photoluminescence, and strain mapping

Single photon emitters (SPEs) in solids have emerged as promising candidates for quantum photonic sensing, communications, and computing. Defects in hexagonal boron nitride (hBN) exhibit high-brightness, room-temperature quantum emission, but their large spectral variability and unknown local structure significantly challenge their technological utility. Here, we directly correlate hBN quantum emission with the material's local strain using a combination of photoluminescence (PL), cathodoluminescence (CL) and nano-beam electron diffraction. Across 40 emitters and 15 samples, we observe zero phonon lines(ZPLs) in PL and CL ranging from 540-720 nm. CL mapping reveals that multiple defects and distinct defect species located within an optically-diffraction-limited region can each contribute to the observed PL spectra. Local strain maps indicate that strain is not required to activate the emitters and is not solely responsible for the observed ZPL spectral range. Instead, four distinct defect classes are responsible for the observed emission range. One defect class has ZPLs near 615 nm with predominantly matched CL-PL responses; it is not a strain-tuned version of another defect class with ZPL emission centered at 580 nm. A third defect class at 650 nm has low visible-frequency CL emission; and a fourth defect species centered at 705 nm has a small, ~10 nm shift between its CL and PL peaks. All studied defects are stable upon both electron and optical irradiation. Our results provide an important foundation for atomic-scale optical characterization of color centers, as well as a foundation for engineering defects with precise emission properties.

cond-mat.mtrl-sci↗

arXiv2018

Effects of microstructure and growth conditions on quantum emitters in Gallium Nitride

Single-photon emitters in gallium nitride (GaN) are gaining interest as attractive quantum systems due to the well-established techniques for growth and nanofabrication of the host material, as well as its remarkable chemical stability and optoelectronic properties. We investigate the nature of such single-photon emitters in GaN with a systematic analysis of various samples produced under different growth conditions. We explore the effect that intrinsic structural defects (dislocations and stacking faults), doping and crystal orientation in GaN have on the formation of quantum emitters. We investigate the relationship between the position of the emitters (determined via spectroscopy and photoluminescence measurements) and the location of threading dislocations (characterised both via atomic force microscopy and cathodoluminescence). We find that quantum emitters do not correlate with stacking faults or dislocations; instead, they are more likely to originate from point defects or impurities whose density is modulated by the local extended defect density.

physics.app-ph↗

arXiv2018

Phonology-Augmented Statistical Framework for Machine Transliteration using Limited Linguistic Resources

Transliteration converts words in a source language (e.g., English) into words in a target language (e.g., Vietnamese). This conversion considers the phonological structure of the target language, as the transliterated output needs to be pronounceable in the target language. For example, a word in Vietnamese that begins with a consonant cluster is phonologically invalid and thus would be an incorrect output of a transliteration system. Most statistical transliteration approaches, albeit being widely adopted, do not explicitly model the target language's phonology, which often results in invalid outputs. The problem is compounded by the limited linguistic resources available when converting foreign words to transliterated words in the target language. In this work, we present a phonology-augmented statistical framework suitable for transliteration, especially when only limited linguistic resources are available. We propose the concept of pseudo-syllables as structures representing how segments of a foreign word are organized according to the syllables of the target language's phonology. We performed transliteration experiments on Vietnamese and Cantonese. We show that the proposed framework outperforms the statistical baseline by up to 44.68% relative, when there are limited training examples (587 entries).

cs.CL↗

arXiv2018

Multimodal neural pronunciation modeling for spoken languages with logographic origin

Graphemes of most languages encode pronunciation, though some are more explicit than others. Languages like Spanish have a straightforward mapping between its graphemes and phonemes, while this mapping is more convoluted for languages like English. Spoken languages such as Cantonese present even more challenges in pronunciation modeling: (1) they do not have a standard written form, (2) the closest graphemic origins are logographic Han characters, of which only a subset of these logographic characters implicitly encodes pronunciation. In this work, we propose a multimodal approach to predict the pronunciation of Cantonese logographic characters, using neural networks with a geometric representation of logographs and pronunciation of cognates in historically related languages. The proposed framework improves performance by 18.1% and 25.0% respective to unimodal and multimodal baselines.

cs.CL↗

arXiv2018

Who is Killed by Police: Introducing Supervised Attention for Hierarchical LSTMs

Finding names of people killed by police has become increasingly important as police shootings get more and more public attention (police killing detection). Unfortunately, there has been not much work in the literature addressing this problem. The early work in this field \cite{keith2017identifying} proposed a distant supervision framework based on Expectation Maximization (EM) to deal with the multiple appearances of the names in documents. However, such EM-based framework cannot take full advantages of deep learning models, necessitating the use of hand-designed features to improve the detection performance. In this work, we present a novel deep learning method to solve the problem of police killing recognition. The proposed method relies on hierarchical LSTMs to model the multiple sentences that contain the person names of interests, and introduce supervised attention mechanisms based on semantical word lists and dependency trees to upweight the important contextual words. Our experiments demonstrate the benefits of the proposed model and yield the state-of-the-art performance for police killing detection.

cs.CL↗

arXiv2018

Bottom up engineering of near-identical quantum emitters in atomically thin materials

Quantum technologies require robust and photostable single photon emitters (SPEs) that can be reliably engineered. Hexagonal boron nitride (hBN) has recently emerged as a promising candidate host to bright and optically stable SPEs operating at room temperature. However, the emission wavelength of the fluorescent defects in hBN has, to date, been shown to be uncontrolled. The emitters usually display a large spread of zero phonon line (ZPL) energies spanning over a broad spectral range (hundreds of nanometers), which hinders the potential development of hBN-based devices and applications. We demonstrate bottom-up, chemical vapor deposition growth of large-area, few layer hBN that hosts large quantities of SPEs: 100 per 10x10 μm2. Remarkably, more than 85 percent of the emitters have a ZPL at (580{\pm}10)nm, a distribution which is over an order of magnitude narrower than previously reported. Exploiting the high density and uniformity of the emitters, we demonstrate electrical modulation and tuning of the ZPL emission wavelength by up to 15 nm. Our results constitute a definite advancement towards the practical deployment of hBN single photon emitters in scalable quantum photonic and hybrid optoelectronic devices based on 2D materials.

physics.app-ph↗

arXiv2018

A Deep Learning Model with Hierarchical LSTMs and Supervised Attention for Anti-Phishing

Anti-phishing aims to detect phishing content/documents in a pool of textual data. This is an important problem in cybersecurity that can help to guard users from fraudulent information. Natural language processing (NLP) offers a natural solution for this problem as it is capable of analyzing the textual content to perform intelligent recognition. In this work, we investigate state-of-the-art techniques for text categorization in NLP to address the problem of anti-phishing for emails (i.e, predicting if an email is phishing or not). These techniques are based on deep learning models that have attracted much attention from the community recently. In particular, we present a framework with hierarchical long short-term memory networks (H-LSTMs) and attention mechanisms to model the emails simultaneously at the word and the sentence level. Our expectation is to produce an effective model for anti-phishing and demonstrate the effectiveness of deep learning for problems in cybersecurity.

cs.CR↗

arXiv2017

Irredundant generating sets and dimension-like invariants of the finite group

Whiston proved that the maximum size of an irredundant generating set in the symmetric group $S_n$ is $n-1$, and Cameron and Cara characterized all irredundant generating sets of $S_n$ that achieve this size. Our goal is to extend their results. Using properties of transitive subgroups of the symmetric group, we are able to classify all irredundant generating sets with sizes $n-2$ in both $A_n$ and $S_n$. Next, based on this classification, we derive other interesting properties for the alternating group $A_n$. Finally, using Whiston's lemma, we will derive the formulas for calculating dimension-like invariants of some specific types of wreath products.

math.GR↗

arXiv2017

Ideals in the Goldman Algebra

The goal of this work is to study the ideals of the Goldman Lie algebra $S$. To do so, we construct an algebra homomorphism from $S$ to a simpler algebraic structure, and focus on finding ideals of this new structure instead. The structure $S$ can be regarded as either a $\mathbb{Q}$-module or a $\mathbb{Q}$-module generated by free homotopy classes. For $\mathbb{Z}$-module case, we proved that there is an infinite class of ideals of $S$ that contain a certain finite set of free homotopy classes. For $\mathbb{Q}$-module case, we can classify all the ideals of the new structure and consequently obtain a new class of ideals of the original structure. Finally, we show an interesting infinite chain of ideals that are not those ideals obtained by considering the new structure.

math.AT↗

arXiv2017

Single Photon Emission from Plasma Treated 2D Hexagonal Boron Nitride

Artificial atomic systems in solids are becoming increasingly important building blocks in quantum information processing and scalable quantum nanophotonic networks. Yet, synthesis of color centers that act as single photon emitters which are suitable for on-chip applications is still beyond reach. Here, we report a number of plasma and thermal annealing methods for the fabrication of emitters in tape-exfoliated hexagonal boron nitride (hBN) crystals. A two-step process comprised of Ar plasma etching and subsequent annealing in Ar is highly robust, and yields a seven-fold increase in the concentration of emitters in hBN. The initial plasma etching step generates emitters that suffer from blinking and bleaching, whereas the two-step process yields emitters that are photostable at room temperature and have an emission energy distribution that is red-shifted relative to that of pristine hBN. An analysis of emitters fabricated by a range of plasma and annealing treatments, combined with a theoretical investigation of point defects in hBN indicates that single photon emitters characterized by a high degree of photostability and emission wavelengths greater than ~700 nm are associated with defect complexes that contain oxygen. This is further confirmed by generating the emitters by annealing hBN in an oxidative atmosphere. Our findings advance present understanding of the structure of quantum emitter in hBN and enhance the nanofabrication toolkit that is needed to realize integrated quantum nanophotonics based on 2D materials.

cond-mat.mtrl-sci↗