arXiv subjects

Ngoc Duy Nguyen

Publications and source records attributed to Ngoc Duy Nguyen.

17 recordsLinked to original sources

arXiv2026

Emissivity by Design: Geometric Mesh Inductance Governs Thermal Radiation in Metallic Nanowire Networks

The thermal infrared emissivity of metallic nanowire networks - the property that makes them candidate low-emissivity transparent electrodes - follows regularities that have resisted physical explanation: a strong wire-diameter dependence with no wire-length dependence, a correlation with sheet resistance that breaks down above ~15 ohm/sq, and an angular signature that turns from dielectric-like to metallic as the network densifies. We show that all of them follow from one local geometric length, the average inter-wire gap g, acting through the geometric inductance of a mesh finer than the thermal wavelength: $X_L = Z_0 (g/λ) \ln(2g/πD)$, a parameter-free reactance that exceeds the ohmic loss and, unlike the DC sheet resistance, stays finite instead of diverging at the percolation threshold. Since the gap is fixed by areal density and wire diameter alone, the optics is length-independent and decoupled from percolative transport; what has been read as a second, optical percolation is instead a smooth impedance crossing. With a single fitted parameter the theory reproduces 56 silver-nanowire samples across four diameters to a mean absolute error of 0.04, collapses them onto one universal curve, and - unchanged - predicts the emissivity-transmittance data of five independent groups and the measured spectral emissivity, including the silica phonon band. Because this reactance contains no material constant, the inter-wire gap emerges as the master design variable for the radiative properties of metallic nanowire transparent conductors.

cond-mat.mes-hall↗

arXiv2026

Challenges and mitigation pathways in coating silver nanowire networks with metallic oxides by RF magnetron sputtering

As silver nanowire (AgNW) networks reach increasing technological maturity, research efforts are progressively shifting toward their integration into functional devices. In this context, it is essential to assess how thin film coating processes affect the structural and functional integrity of these transparent conducting networks. Radio Frequency (RF) magnetron sputtering is among the most widely used and industrially scalable deposition techniques, making a detailed understanding of its impact on AgNW networks particularly critical. In this work, we experimentally investigate the degradation of AgNW networks observed under specific RF magnetron sputtering regimes. By varying deposition time, oxygen partial pressure, target material, buffer layers and plasma power, we analyze how sputtering conditions influence the electrical, morphological, and structural properties of the networks. Based on these observations, we identify viable strategies to mitigate or suppress network degradation, thereby enabling safer and more reliable coating protocols. These results provide practical guidelines for the integration of AgNW networks into multilayer device architectures.

cond-mat.mtrl-sci↗

arXiv2022

vieCap4H-VLSP 2021: Vietnamese Image Captioning for Healthcare Domain using Swin Transformer and Attention-based LSTM

This study presents our approach on the automatic Vietnamese image captioning for healthcare domain in text processing tasks of Vietnamese Language and Speech Processing (VLSP) Challenge 2021, as shown in Figure 1. In recent years, image captioning often employs a convolutional neural network-based architecture as an encoder and a long short-term memory (LSTM) as a decoder to generate sentences. These models perform remarkably well in different datasets. Our proposed model also has an encoder and a decoder, but we instead use a Swin Transformer in the encoder, and a LSTM combined with an attention module in the decoder. The study presents our training experiments and techniques used during the competition. Our model achieves a BLEU4 score of 0.293 on the vietCap4H dataset, and the score is ranked the 3$^{rd}$ place on the private leaderboard. Our code can be found at \url{https://git.io/JDdJm}.

cs.CV↗

arXiv2021

Relevance of Ge incorporation to control the physical behaviour of point defects in kesterite

To reduce the prominent VOC-deficit that limits kesterite-based solar cells efficiencies, Ge has been proposed over the recent years with encouraging results, as the reduction of the non-radiative recombination rate is considered as a way to improve the well-known Sn-kesterite world record efficiency. To gain further insight into this mechanism, we investigate the physical behaviour of intrinsic point defects both upon Ge doping and alloying of Cu2ZnSnS4 kesterite. Using a first-principles approach, we confirm the p-type conductivity of both Cu2ZnSnS4 and Cu2ZnGeS4, attributed to the low formation energies of the VCu and CuZn acceptor defects within the whole stable phase diagram range. Via doping of the Sn-kesterite matrix, we report the lowest formation energy for the substitutional defect GeSn. We also confirm the detrimental role of the substitutional defects XZn (X=Sn,Ge) acting as recombination centres within the Sn-based, the Ge-doped and the Ge-based kesterite. Finally, we highlight the reduction of the lattice distortion upon Ge incorporation resulting in a reduction of the carrier capture cross section and consequently a decrease of the non-radiative recombination rate within the bulk material.

cond-mat.mtrl-sci↗

arXiv2021

Fruit-CoV: An Efficient Vision-based Framework for Speedy Detection and Diagnosis of SARS-CoV-2 Infections Through Recorded Cough Sounds

SARS-CoV-2 is colloquially known as COVID-19 that had an initial outbreak in December 2019. The deadly virus has spread across the world, taking part in the global pandemic disease since March 2020. In addition, a recent variant of SARS-CoV-2 named Delta is intractably contagious and responsible for more than four million deaths over the world. Therefore, it is vital to possess a self-testing service of SARS-CoV-2 at home. In this study, we introduce Fruit-CoV, a two-stage vision framework, which is capable of detecting SARS-CoV-2 infections through recorded cough sounds. Specifically, we convert sounds into Log-Mel Spectrograms and use the EfficientNet-V2 network to extract its visual features in the first stage. In the second stage, we use 14 convolutional layers extracted from the large-scale Pretrained Audio Neural Networks for audio pattern recognition (PANNs) and the Wavegram-Log-Mel-CNN to aggregate feature representations of the Log-Mel Spectrograms. Finally, we use the combined features to train a binary classifier. In this study, we use a dataset provided by the AICovidVN 115M Challenge, which includes a total of 7371 recorded cough sounds collected throughout Vietnam, India, and Switzerland. Experimental results show that our proposed model achieves an AUC score of 92.8% and ranks the 1st place on the leaderboard of the AICovidVN Challenge. More importantly, our proposed framework can be integrated into a call center or a VoIP system to speed up detecting SARS-CoV-2 infections through online/recorded cough sounds.

cs.SD↗

arXiv2021

A Visual Communication Map for Multi-Agent Deep Reinforcement Learning

Deep reinforcement learning has been applied successfully to solve various real-world problems and the number of its applications in the multi-agent settings has been increasing. Multi-agent learning distinctly poses significant challenges in the effort to allocate a concealed communication medium. Agents receive thorough knowledge from the medium to determine subsequent actions in a distributed nature. Apparently, the goal is to leverage the cooperation of multiple agents to achieve a designated objective efficiently. Recent studies typically combine a specialized neural network with reinforcement learning to enable communication between agents. This approach, however, limits the number of agents or necessitates the homogeneity of the system. In this paper, we have proposed a more scalable approach that not only deals with a great number of agents but also enables collaboration between dissimilar functional agents and compatibly combined with any deep reinforcement learning methods. Specifically, we create a global communication map to represent the status of each agent in the system visually. The visual map and the environmental state are fed to a shared-parameter network to train multiple agents concurrently. Finally, we select the Asynchronous Advantage Actor-Critic (A3C) algorithm to demonstrate our proposed scheme, namely Visual communication map for Multi-agent A3C (VMA3C). Simulation results show that the use of visual communication map improves the performance of A3C regarding learning speed, reward achievement, and robustness in multi-agent problems.

cs.LG↗

arXiv2021

Review, Analysis and Design of a Comprehensive Deep Reinforcement Learning Framework

The integration of deep learning to reinforcement learning (RL) has enabled RL to perform efficiently in high-dimensional environments. Deep RL methods have been applied to solve many complex real-world problems in recent years. However, development of a deep RL-based system is challenging because of various issues such as the selection of a suitable deep RL algorithm, its network configuration, training time, training methods, and so on. This paper proposes a comprehensive software framework that not only plays a vital role in designing a connect-the-dots deep RL architecture but also provides a guideline to develop a realistic RL application in a short time span. We have designed and developed a deep RL-based software framework that strictly ensures flexibility, robustness, and scalability. By inheriting the proposed architecture, software managers can foresee any challenges when designing a deep RL-based system. As a result, they can expedite the design process and actively control every stage of software development, which is especially critical in agile development environments. To enforce generalization, the proposed architecture does not depend on a specific RL algorithm, a network configuration, the number of agents, or the type of agents. Using our framework, software developers can develop and integrate new RL algorithms or new types of agents, and can flexibly change network configuration or the number of agents.

cs.LG↗

arXiv2020

Opto-electronic properties and solar cell efficiency modelling of Cu$_2$ZnXS$_4$ (X=Sn,Ge,Si) kesterites

In this work, first principle calculations of Cu$_2$ZnSnS$_4$ (CZTS), Cu$_2$ZnGeS$_4$ (CZGS) and Cu$_2$ZnSiS$_4$ (CZSS) are performed to highlight the impact of the cationic substitution on the structural, electronic and optical properties of kesterite compounds. Direct bandgaps are reported with values of 1.32, 1.89 and 3.06 eV respectively for CZTS, CZGS and CZSS. In addition, absorption coefficient values of the order of $10^4$ cm$^{-1}$ are obtained, indicating the applicability of these materials as absorber layer for solar cell applications. In the second part of this study, ab initio results are used as input data to model the electrical power conversion efficiency of kesterite-based solar cell. In that perspective, we used an improved version of the Shockley-Queisser theoretical model including non-radiative recombination via an external parameter defined as the internal quantum efficiency. Based on predicted optimal absorber layer thicknesses, the variation of the solar cell maximal efficiency is studied as a function of the non-radiative recombination rate. Maximal efficiencies of 25.88, 19.94 and 3.11% are reported respectively for CZTS, CZGS and CZSS for vanishing non-radiative recombination rate. Using an internal quantum efficiency providing $V_{OC}$ values comparable to experimental measurements, solar cell efficiencies of 15.88, 14.98 and 2.66% are reported respectively for CZTS, CZGS and CZSS (for an optimal thickness of 1.15 $μ$m). With this methodology, we confirm the suitability of CZTS in single junction solar cells, with a possible efficiency improvement of 10% enabled through the reduction of the non-radiative recombination rate. In addition, CZGS appears to be an interesting candidate as top cell absorber layer for tandem approaches whereas CZSS might be interesting for transparent PV windows.

physics.app-ph↗

arXiv2020

A Multi-Objective Deep Reinforcement Learning Framework

This paper introduces a new scalable multi-objective deep reinforcement learning (MODRL) framework based on deep Q-networks. We develop a high-performance MODRL framework that supports both single-policy and multi-policy strategies, as well as both linear and non-linear approaches to action selection. The experimental results on two benchmark problems (two-objective deep sea treasure environment and three-objective Mountain Car problem) indicate that the proposed framework is able to find the Pareto-optimal solutions effectively. The proposed framework is generic and highly modularized, which allows the integration of different deep reinforcement learning algorithms in different complex problem domains. This therefore overcomes many disadvantages involved with standard multi-objective reinforcement learning methods in the current literature. The proposed framework acts as a testbed platform that accelerates the development of MODRL for solving increasingly complicated multi-objective problems.

cs.LG↗

arXiv2020

On the origin of the giant spin detection efficiency in tunnel barrier based electrical spin detector

Efficient conversion of a spin signal into an electric voltage in mainstream semiconductors is one of the grand challenges of spintronics. This process is commonly achieved via a ferromagnetic tunnel barrier where non-linear electric transport occurs. In this work, we demonstrate that non-linearity may lead to a spin-to-charge conversion efficiency larger than 10 times the spin polarization of the tunnel barrier when the latter is under bias of a few mV. We identify the underlying mechanisms responsible for this remarkably efficient spin detection as the tunnel barrier deformation and the conduction band shift resulting from a change of applied voltage. In addition, we derive an approximate analytical expression for the detector spin sensitivity $P_{\textrm{det}}(V)$. Calculations performed for different barrier shapes show that this enhancement is present in oxide barriers as well as in Schottky tunnel barriers even if the dominant mechanisms differs with the barrier type. Moreover, although the spin signal is reduced at high temperatures, it remains superior to the value predicted by the linear model. Our findings shed light into the interpretation and understanding of electrical spin detection experiments and open new paths to optimize the performance of spin transport devices.

physics.app-ph↗

arXiv2019

Trapping of electrons around nanoscale metallic wires embedded in a semiconductor medium

We predict that conduction electrons in a semiconductor film containing a centered square array of metal nanowires normal to its plane are bound in quantum states around the central wires, if a positive bias voltage is applied between the wires at the square vertices and these latter. We obtain and discuss the eigenenergies and eigenfunctions of two models with different dimensions. The results show that the eigenstates can be grouped into different shells. The energy differences between the shells is typically a few tens of meV, which corresponds to frequencies of emitted or absorbed photons in a range of 3 THz to 20 THz approximately. These energy differences strongly depend on the bias voltage. We calculate the linear response of individual electrons on the ground level of our models to large-wavelength electromagnetic waves whose electric field is in the plane of the semiconductor film. The computed oscillator strengths are dominated by the transitions to the states in each shell whose wave function has a single radial node line normal to the wave electric field. We include the effect of the image charge induced on the central metal wires and show that it modifies the oscillator strengths so that their sum deviates from the value given by the Thomas-Reiche-Kuhn rule. We report the linear response, or polarizability, versus photon energy, of the studied models and their absorption spectra. These latter show well-defined peaks as expected from the study of the oscillator strengths. We show that the position of these absorption peaks is strongly dependent on the bias voltage so that the frequency of photon absorption or emission in the systems described here is easily tunable. This makes them good candidates for the development of novel infrared devices.

cond-mat.mes-hall↗

arXiv2019

A Review of Situation Awareness Assessment Approaches in Aviation Environments

Situation awareness (SA) is an important constituent in human information processing and essential in pilots' decision-making processes. Acquiring and maintaining appropriate levels of SA is critical in aviation environments as it affects all decisions and actions taking place in flights and air traffic control. This paper provides an overview of recent measurement models and approaches to establishing and enhancing SA in aviation environments. Many aspects of SA are examined including the classification of SA techniques into six categories, and different theoretical SA models from individual, to shared or team, and to distributed or system levels. Quantitative and qualitative perspectives pertaining to SA methods and issues of SA for unmanned vehicles are also addressed. Furthermore, future research directions regarding SA assessment approaches are raised to deal with shortcomings of the existing state-of-the-art methods in the literature.

cs.HC↗

arXiv2019

Multi-Agent Deep Reinforcement Learning with Human Strategies

Deep learning has enabled traditional reinforcement learning methods to deal with high-dimensional problems. However, one of the disadvantages of deep reinforcement learning methods is the limited exploration capacity of learning agents. In this paper, we introduce an approach that integrates human strategies to increase the exploration capacity of multiple deep reinforcement learning agents. We also report the development of our own multi-agent environment called Multiple Tank Defence to simulate the proposed approach. The results show the significant performance improvement of multiple agents that have learned cooperatively with human strategies. This implies that there is a critical need for human intellect teamed with machines to solve complex problems. In addition, the success of this simulation indicates that our multi-agent environment can be used as a testbed platform to develop and validate other multi-agent control algorithms.

cs.LG↗

arXiv2019

Manipulating Soft Tissues by Deep Reinforcement Learning for Autonomous Robotic Surgery

In robotic surgery, pattern cutting through a deformable material is a challenging research field. The cutting procedure requires a robot to concurrently manipulate a scissor and a gripper to cut through a predefined contour trajectory on the deformable sheet. The gripper ensures the cutting accuracy by nailing a point on the sheet and continuously tensioning the pinch point to different directions while the scissor is in action. The goal is to find a pinch point and a corresponding tensioning policy to minimize damage to the material and increase cutting accuracy measured by the symmetric difference between the predefined contour and the cut contour. Previous study considers finding one fixed pinch point during the course of cutting, which is inaccurate and unsafe when the contour trajectory is complex. In this paper, we examine the soft tissue cutting task by using multiple pinch points, which imitates human operations while cutting. This approach, however, does not require the use of a multi-gripper robot. We use a deep reinforcement learning algorithm to find an optimal tensioning policy of a pinch point. Simulation results show that the multi-point approach outperforms the state-of-the-art method in soft pattern cutting task with respect to both accuracy and reliability.

cs.RO↗

arXiv2019

Deep Reinforcement Learning for Multi-Agent Systems: A Review of Challenges, Solutions and Applications

Reinforcement learning (RL) algorithms have been around for decades and employed to solve various sequential decision-making problems. These algorithms however have faced great challenges when dealing with high-dimensional environments. The recent development of deep learning has enabled RL methods to drive optimal policies for sophisticated and capable agents, which can perform efficiently in these challenging environments. This paper addresses an important aspect of deep RL related to situations that require multiple agents to communicate and cooperate to solve complex tasks. A survey of different approaches to problems related to multi-agent deep RL (MADRL) is presented, including non-stationarity, partial observability, continuous state and action spaces, multi-agent training schemes, multi-agent transfer learning. The merits and demerits of the reviewed methods will be analyzed and discussed, with their corresponding applications explored. It is envisaged that this review provides insights about various MADRL methods and can lead to future development of more robust and highly useful multi-agent learning methods for solving real-world problems.

cs.LG↗

arXiv2019

A New Tensioning Method using Deep Reinforcement Learning for Surgical Pattern Cutting

Surgeons normally need surgical scissors and tissue grippers to cut through a deformable surgical tissue. The cutting accuracy depends on the skills to manipulate these two tools. Such skills are part of basic surgical skills training as in the Fundamentals of Laparoscopic Surgery. The gripper is used to pinch a point on the surgical sheet and pull the tissue to a certain direction to maintain the tension while the scissors cut through a trajectory. As the surgical materials are deformable, it requires a comprehensive tensioning policy to yield appropriate tensioning direction at each step of the cutting process. Automating a tensioning policy for a given cutting trajectory will support not only the human surgeons but also the surgical robots to improve the cutting accuracy and reliability. This paper presents a multiple pinch point approach to modelling an autonomous tensioning planner based on a deep reinforcement learning algorithm. Experiments on a simulator show that the proposed method is superior to existing methods in terms of both performance and robustness.

cs.RO↗

arXiv2018

A Human Mixed Strategy Approach to Deep Reinforcement Learning

In 2015, Google's DeepMind announced an advancement in creating an autonomous agent based on deep reinforcement learning (DRL) that could beat a professional player in a series of 49 Atari games. However, the current manifestation of DRL is still immature, and has significant drawbacks. One of DRL's imperfections is its lack of "exploration" during the training process, especially when working with high-dimensional problems. In this paper, we propose a mixed strategy approach that mimics behaviors of human when interacting with environment, and create a "thinking" agent that allows for more efficient exploration in the DRL training process. The simulation results based on the Breakout game show that our scheme achieves a higher probability of obtaining a maximum score than does the baseline DRL algorithm, i.e., the asynchronous advantage actor-critic method. The proposed scheme therefore can be applied effectively to solving a complicated task in a real-world application.

cs.LG↗