arXiv subjects

Tri Nguyen

Publications and source records attributed to Tri Nguyen.

65 records · Page 4Linked to original sources

arXiv2021

Applications and Techniques for Fast Machine Learning in Science

In this community review report, we discuss applications and techniques for fast machine learning (ML) in science -- the concept of integrating power ML methods into the real-time experimental data processing loop to accelerate scientific discovery. The material for the report builds on two workshops held by the Fast ML for Science community and covers three main areas: applications for fast ML across a number of scientific domains; techniques for training and implementing performant and resource-efficient ML algorithms; and computing architectures, platforms, and technologies for deploying these algorithms. We also present overlapping challenges across the multiple scientific domains where common solutions can be found. This community report is intended to give plenty of examples and inspiration for scientific discovery through integrated and accelerated ML solutions. This is followed by a high-level overview and organization of technical advances, including an abundance of pointers to source material, which can enable these breakthroughs.

cs.LG↗

arXiv2021

A Novel Internet-of-Drones and Blockchain-based System Architecture for Search and Rescue

With the development in information and communications technology (ICT) and drones such as Internet-of-Things (IoT), edge computing, image processing, and autonomous drones, solutions supporting search and rescue (SAR) missions can be developed with more intelligent capabilities. In most of the drone and unmanned aerial vehicle (UAV) based systems supporting SAR missions, several drones deployed in different areas acquire images and videos that are sent to a ground control station (GCS) for processing and detecting a missing person. Although this offers many advantages, such as easy management and deployment, the approach still has many limitations. For example, when a connection between a drone and a GCS has some problems, the quality of service cannot be maintained. Many drone and UAV-based systems do not support flexibility, transparency, security, and traceability. In this paper, we propose a novel Internet-of-Drones (IoD) architecture using blockchain technology. We implement the proposed architecture with different drones, edge servers, and a Hyperledger blockchain network. The proof-of-concept design demonstrates that the proposed architecture can offer high-level services such as prolonging the operating time of a drone, improving the capability of detecting humans accurately, and a high level of transparency, traceability, and security.

cs.NI↗

arXiv2021

Hardware-accelerated Inference for Real-Time Gravitational-Wave Astronomy

The field of transient astronomy has seen a revolution with the first gravitational-wave detections and the arrival of multi-messenger observations they enabled. Transformed by the first detection of binary black hole and binary neutron star mergers, computational demands in gravitational-wave astronomy are expected to grow by at least a factor of two over the next five years as the global network of kilometer-scale interferometers are brought to design sensitivity. With the increase in detector sensitivity, real-time delivery of gravitational-wave alerts will become increasingly important as an enabler of multi-messenger followup. In this work, we report a novel implementation and deployment of deep learning inference for real-time gravitational-wave data denoising and astrophysical source identification. This is accomplished using a generic Inference-as-a-Service model that is capable of adapting to the future needs of gravitational-wave data analysis. Our implementation allows seamless incorporation of hardware accelerators and also enables the use of commercial or private (dedicated) as-a-service computing. Based on our results, we propose a paradigm shift in low-latency and offline computing in gravitational-wave astronomy. Such a shift can address key challenges in peak-usage, scalability and reliability, and provide a data analysis platform particularly optimized for deep learning applications. The achieved sub-millisecond scale latency will also be relevant for any machine learning-based real-time control systems that may be invoked in the operation of near-future and next generation ground-based laser interferometers, as well as the front-end collection, distribution and processing of data from such instruments.

gr-qc↗

arXiv2021

StressNAS: Affect State and Stress Detection Using Neural Architecture Search

Smartwatches have rapidly evolved towards capabilities to accurately capture physiological signals. As an appealing application, stress detection attracts many studies due to its potential benefits to human health. It is propitious to investigate the applicability of deep neural networks (DNN) to enhance human decision-making through physiological signals. However, manually engineering DNN proves a tedious task especially in stress detection due to the complex nature of this phenomenon. To this end, we propose an optimized deep neural network training scheme using neural architecture search merely using wrist-worn data from WESAD. Experiments show that our approach outperforms traditional ML methods by 8.22% and 6.02% in the three-state and two-state classifiers, respectively, using the combination of WESAD wrist signals. Moreover, the proposed method can minimize the need for human-design DNN while improving performance by 4.39% (three-state) and 8.99% (binary).

cs.LG↗

arXiv2020

Glassy Carbon Microelectrode Arrays Enable Voltage-Peak Separated Simultaneous Detection of Dopamine and Serotonin Using Fast Scan Cyclic Voltammetry

Progress in real-time, simultaneous in vivo detection of multiple neurotransmitters will help accelerate advances in neuroscience research. The need for development of probes capable of stable electrochemical detection of rapid neurotransmitter fluctuations with high sensitivity and selectivity and sub-second temporal resolution has, therefore, become compelling. Additionally, a higher spatial resolution multi-channel capability is required to capture the complex neurotransmission dynamics across different brain regions. These research needs have inspired the introduction of glassy carbon (GC) microelectrode arrays on flexible polymer substrates through carbon MEMS (C-MEMS) microfabrication process followed by a novel pattern transfer technique. These implantable GC microelectrodes offer unique advantages in electrochemical detection of electroactive neurotransmitters through the presence of active carboxyl, carbonyl, and hydroxyl functional groups. In addition, they offer fast electron transfer kinetics, capacitive electrochemical behavior, and wide electrochemical window. Here, we combine the use of these GC microelectrodes with the fast scan cyclic voltammetry (FSCV) technique to optimize the co-detection of dopamine and serotonin in vitro and in vivo. We demonstrate that using optimized FSCV triangular waveform at scan rates lower than 700 V/s and holding and switching at potentials of 0.4 and 1V respectively, it is possible to discriminate voltage reduction and oxidation peaks of serotonin and dopamine, with serotonin contributing distinct multiple oxidation peaks. Taken together, our results present a compelling case for a carbon-based MEA platform rich with active functional groups that allows for repeatable and stable detection of electroactive multiple neurotransmitters at concentrations as low as 10 nM

q-bio.BM↗

arXiv2020

Noise Reduction in Gravitational-wave Data via Deep Learning

With the advent of gravitational wave astronomy, techniques to extend the reach of gravitational wave detectors are desired. In addition to the stellar-mass black hole and neutron star mergers already detected, many more are below the surface of the noise, available for detection if the noise is reduced enough. Our method (DeepClean) applies machine learning algorithms to gravitational wave detector data and data from on-site sensors monitoring the instrument to reduce the noise in the time-series due to instrumental artifacts and environmental contamination. This framework is generic enough to subtract linear, non-linear, and non-stationary coupling mechanisms. It may also provide handles in learning about the mechanisms which are not currently understood to be limiting detector sensitivities. The robustness of the noise reduction technique in its ability to efficiently remove noise with no unintended effects on gravitational-wave signals is also addressed through software signal injection and parameter estimation of the recovered signal. It is shown that the optimal SNR ratio of the injected signal is enhanced by $\sim 21.6\%$ and the recovered parameters are consistent with the injected set. We present the performance of this algorithm on linear and non-linear noise sources and discuss its impact on astrophysical searches by gravitational wave detectors.

astro-ph.IM↗

arXiv2020

Enhancing Gravitational-Wave Science with Machine Learning

Machine learning has emerged as a popular and powerful approach for solving problems in astrophysics. We review applications of machine learning techniques for the analysis of ground-based gravitational-wave detector data. Examples include techniques for improving the sensitivity of Advanced LIGO and Advanced Virgo gravitational-wave searches, methods for fast measurements of the astrophysical parameters of gravitational-wave sources, and algorithms for reduction and characterization of non-astrophysical detector noise. These applications demonstrate how machine learning techniques may be harnessed to enhance the science that is possible with current and future gravitational-wave detectors.

astro-ph.HE↗

arXiv2019

Combination of Domain Knowledge and Deep Learning for Sentiment Analysis of Short and Informal Messages on Social Media

Sentiment analysis has been emerging recently as one of the major natural language processing (NLP) tasks in many applications. Especially, as social media channels (e.g. social networks or forums) have become significant sources for brands to observe user opinions about their products, this task is thus increasingly crucial. However, when applied with real data obtained from social media, we notice that there is a high volume of short and informal messages posted by users on those channels. This kind of data makes the existing works suffer from many difficulties to handle, especially ones using deep learning approaches. In this paper, we propose an approach to handle this problem. This work is extended from our previous work, in which we proposed to combine the typical deep learning technique of Convolutional Neural Networks with domain knowledge. The combination is used for acquiring additional training data augmentation and a more reasonable loss function. In this work, we further improve our architecture by various substantial enhancements, including negation-based data augmentation, transfer learning for word embeddings, the combination of word-level embeddings and character-level embeddings, and using multitask learning technique for attaching domain knowledge rules in the learning process. Those enhancements, specifically aiming to handle short and informal messages, help us to enjoy significant improvement in performance once experimenting on real datasets.

cs.CL↗

arXiv2019

Transport Properties of Near Surface InAs Two-dimensional Heterostructures

Two-dimensional electron systems (2DESs) confined to the surface of narrowband semiconductors have attracted great interest since they can easily integrate with superconductivity (or ferromagnetism) enabling new possibilities in hybrid device architectures and study of exotic states in proximity of superconductors. In this work, we study indium arsenide heterostructures where combination of clean interface with superconductivity, high mobility and spin-orbit coupling can be achieved. The weak antilocalization measurements indicate presence of strong spin-orbit coupling at high densities. We study the magnetotransport as a function of top barrier and density and report clear observation of integer quantum Hall states. We report improved electron mobility reaching up to 44,000 cm$^{2}$/Vs in undoped heterstructures and well developed integer quantum Hall states starting as low as 2.5~T.

cond-mat.mes-hall↗

arXiv2019

Superconducting proximity effect in epitaxial Al-InAs heterostructures

Semiconductor-based Josephson junctions provide a platform for studying proximity effect due to the possibility of tuning junction properties by gate voltage and large-scale fabrication of complex Josephson circuits. Recently Josephson junctions using InAs weak link with epitaxial aluminum contact have improved the product of normal resistance and critical current, $I_cR_N$, in addition to fabrication process reliability. Here we study similar devices with epitaxial contact and find large supercurrent and substantial product of $I_cR_N$ in our junctions. However we find a striking difference when we compare these samples with higher mobility samples in terms of product of excess current and normal resistance, $I_{ex}R_N$. The excess current is negligible in lower mobility devices while it is substantial and independent of gate voltage and junction length in high mobility samples. This indicates that even though both sample types have epitaxial contacts only the high-mobility one has a high transparency interface. In the high mobility short junctions, we observe values of $I_cR_N/Δ\sim 2.2$ and $I_{ex}R_N/Δ\sim 1.5$ in semiconductor weak links.

cond-mat.mes-hall↗

arXiv2018

MS MARCO: A Human Generated MAchine Reading COmprehension Dataset

We introduce a large scale MAchine Reading COmprehension dataset, which we name MS MARCO. The dataset comprises of 1,010,916 anonymized questions---sampled from Bing's search query logs---each with a human generated answer and 182,669 completely human rewritten generated answers. In addition, the dataset contains 8,841,823 passages---extracted from 3,563,535 web documents retrieved by Bing---that provide the information necessary for curating the natural language answers. A question in the MS MARCO dataset may have multiple answers or no answers at all. Using this dataset, we propose three different tasks with varying levels of difficulty: (i) predict if a question is answerable given a set of context passages, and extract and synthesize the answer as a human would (ii) generate a well-formed answer (if possible) based on the context passages that can be understood with the question and passage context, and finally (iii) rank a set of retrieved passages given a question. The size of the dataset and the fact that the questions are derived from real user search queries distinguishes MS MARCO from other well-known publicly available datasets for machine reading comprehension and question-answering. We believe that the scale and the real-world nature of this dataset makes it attractive for benchmarking machine reading comprehension and question-answering models.

cs.CL↗