DR-NTU (Digital Repository of NTU)
Not a member yet
116018 research outputs found
Sort by
Pseudo-density operators and tests for quantum gravity
No full textThe pseudo-density operators formalism was proposed to describe spatial and temporal
(quantum) correlations on an equal footing. Studying correlations arising in quantum systems
has a history as long as the field of quantum information since the proposal of Bell’s
inequalities and studying them could unveil the very nature of the physical world. In an
effort to extend the formalism of pseudo-density operators, we proposed a description of
pseudo-density operators which deals with discrete and continuous quantum observables
simultaneously and a physically sensible procedure to carry out post-selection in the formalism.
We then made an attempt to describe the Bose-Marletto-Vedral experimental proposal
in the pseudo-density operator formalism, which yielded a modified version of the proposal.
We demonstrated using a causal measure of the pseudo-density operator of the modified
system that temporal correlations are mediated by a linear quantum gravity field between
two quantum probes and proposed that the same causal measure could be used to witness
the non-classicality of gravity by rejecting the Newton-Schroedinger model in favour of linear
quantum gravity, thus showing the benefit of treating quantum correlations in the pseudodensity
operator formalism.Bachelor's degre
Glass transition in monolayers of rough colloidal ellipsoids
No full textStructure-dynamics correlation is one of the major ongoing debates in the glass transition, although a number of structural features have been found connected to the dynamic heterogeneity in different glass-forming colloidal systems. Here, using colloidal experiments combined with coarse-grained molecular dynamics simulations, we investigate the glass transition in monolayers of rough colloidal ellipsoids. Compared with smooth colloidal ellipsoids, the surface roughness of ellipsoids is found to significantly change the nature of glass transition. In particular, we find that the surface roughness induced by coating only a few small hemispheres on the ellipsoids can eliminate the existence of orientational glass and the two-step glass transition found in monolayers of smooth ellipsoids. This is due to the surface roughness-induced coupling between the translational and rotational degrees of freedom in colloidal ellipsoids, which also destroys the structure-dynamics correlation found in glass-forming suspensions of colloidal ellipsoids. Our results not only suggest a new way of using surface roughness to manipulate the glass transition in colloidal systems, but also highlight the importance of detailed particle shape on the glass transition and structure-dynamics correlation in suspensions of anisotropic colloids.Ministry of Education (MOE)National Research Foundation (NRF)This work was financially supported by the National Natural Science Foundation of China (12074275, 12474196, and 11704269), the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (20KJA150008 and 17KJB140020), the PAPD program of Jiangsu Higher Education Institutions, the State and Local Joint Engineering Laboratory for Novel Functional Polymeric Materials, the Jiangsu Key Laboratory of Advanced Functional Polymer Materials, the Jiangsu Engineering Laboratory of Novel Functional Polymeric Materials, the Space Application System of China Manned Space Program (KJZ-YY-NLT0501), the Academic Research Fund from the Singapore Ministry of Education (RG151/23 and MOE2019-T2-2-010), and the National Research Foundation, Singapore, under its 29th Competitive Research Program (CRP) Call (NRF-CRP29-2022-0002)
Approximate computing for machine learning
No full textApproximate Computing has emerged as a promising paradigm to enhance
computational efficiency by introducing controlled inaccuracies in arithmetic
operations. This study explores the application of static approximate adders (AAs)
within a machine learning context, specifically the K-means clustering algorithm, to
assess the trade-offs between clustering accuracy and computational resource
savings. A total of 17 AAs, including two newly proposed designs (BPAA-LSP1 and NAA),
along with a conventional accurate adder, were integrated into a K-means clustering
algorithm and evaluated across four benchmark datasets. Both software simulations
and hardware implementations were conducted, measuring key performance metrics
such as within-cluster sum of squares (WCSS), power, delay, and area.
Results demonstrate that approximate adders maintain strong clustering performance
while significantly reducing power consumption (up to 60% lower power-delay product)
and area usage (up to 50% lower area-delay product). Notably, the newly proposed
BPAA-LSP1 and NAA adders achieve an outstanding balance between clustering
accuracy and computational efficiency, comparable to those achieved using the
accurate adder. BPAA-LSP1 demonstrates reductions of 22.2% in power, 21.6% in area,
and 26.3% in delay, while NAA achieves even greater efficiency, with 31.0% lower
power consumption, 21.6% reduced area, and a 37.1% decrease in delay.
These results suggest that specific characteristics of the error generated by these AAs
may be beneficial or have error-compensatory effects for iterative machine learning
tasks. These findings indicate that approximate adders, particularly BPAA-LSP1 and
NAA, could serve as viable alternatives in energy-efficient, error-tolerant machine
learning applications.Bachelor's degre
Statistical and financial applications of constrained Dantzig-type estimators
No full textTraditional optimisation techniques face significant challenges in handling high-dimensional data with limited and noisy information. This paper explores the use of the Constrained Dantzig-type Estimator (CDE) in two distinct scenarios. First, to optimise website advertising by adapting a technique previously applied in sparse portfolio construction to the domain of online advertising. Unlike prior methods such as Constrained Lasso (CLasso), CDE allows the problem to be formulated as a Linear Programming Problem (LP) instead of a Quadratic one, offering substantial computational advantages in high-dimensional settings while encouraging sparsity in its solutions. In a high-dimensional setting, where the sample covariance is often singular, we explore three different formulations of CDE to estimate the precision matrix as an alternative to existing methods to produce a sparse and positive-definite solution while directly incorporating symmetric conditions into the optimisation problem, comparing their performances to a previously tested benchmark, the Constrained -Minimization for Inverse Matrix Estimation (CLIME).
Finally, we provide a theoretical analysis of CDE V3 by recasting the matrix problem as a single -dimensional CDE. Under mild regularity, sparsity, and restricted eigenvalue (RE) conditions, we prove non-asymptotic and Frobenius-norm error bounds of order , where is the number of nonzeros and is the RE constant.Bachelor's degre
Propionic acid enhances H2 production in purple phototrophic bacteria: insights into carbon and reducing equivalent allocation
No full textBiohydrogen is gaining popularity as a clean and cost-effective energy source. Among the various production methods, photo fermentation (PF) with purple phototrophic bacteria (PPB) has shown great opportunity due to its high hydrogen yield. In practice, this yield is influenced by several factors, with the carbon source, particularly simple organic acid, being a key element that has attracted considerable research interest. Short-chain volatile fatty acids (VFAs), such as acetate, propionate, and butyrate, are widely found in waste streams and dark fermentation (DF) effluent. However, most studies on these VFAs focus mainly on performance evaluation, with few exploring the underlying mechanisms, which limits their applicability in real-world scenarios. To uncover the metabolic mechanisms, this study uses metagenomics to clarify the processes of reducing power production and distribution during substrate assimilation. Meanwhile, this study presents the impact of short-chain VFAs on biohydrogen, polyhydroxyalkanoates (PHA) and glycogen production by PPB. The results show that: (1) over long-term cultivation at similar COD consumption rates of 0.06 g COD/d, PPB possessed the highest hydrogen yield when fed with propionate (0.620 L H2·g COD-1) compared with butyrate (0.434) and acetate (0.361); (2) with propionate as the substrate, PPB accumulated less PHA (7 % of dry biomass) but more glycogen content (11 %), compared to butyrate (15 % PHA and 8 % glycogen) and acetate (21 % PHA and 5 % glycogen); (3) metagenomic analysis revealed that propionate resulted in the highest amounts of reducing equivalents, followed by butyrate and acetate; hydrogen production was the most efficient pathway for utilizing the reducing power with propionate, as the CO2 fixation and PHA or glycogen synthesis were ineffective for electron dissipation. This study offers insights into metabolic mechanism that could guide waste stream selection and pretreatment processes to provide favorable VFAs for the PF process, thereby enhancing PPB biohydrogen production performance in practical applications.Agency for Science, Technology and Research (A*STAR)Ministry of Education (MOE)Nanyang Technological UniversityThe authors would like to thank the Nanyang Environment & Water Research Institute and the Interdisciplinary Graduate Programme, Nanyang Technological University, Singapore, for the award of research scholarship. This work was supported by A*STAR SFS IAF-PP grant (A20H7a0152) and AcRF Tier 1 RT 12/20 awarded to Yan Zhou
Student case management system: a web-based dashboard for NTU staff
No full textThis report presents the development of a comprehensive web application designed for
Nanyang Technological University (NTU) staff to efficiently monitor and manage student
cases. The system enables staff to create, retrieve, update, and delete cases across five
major categories: Leave of Absence, Withdrawal, Discipline, Special Education Needs,
and Pastoral matters. The application features a dynamic dashboard displaying statis-
tics, comprehensive pages listing all students with active cases, and specialized category
pages for focused case management. Each module incorporates advanced functionalities
including bulk case creation through CSV import/export capabilities and intuitive data
visualization. Built using modern web technologies including TypeScript, React with Re-
fine framework, and GraphQL, the system provides a scalable, secure, and user-friendly
solution that significantly streamlines the university’s case management processes.Bachelor's degre
Classification of cognitive syndromes in a Southeast Asian population: interpretable graph convolutional neural networks
No full textDementia is a debilitating disease that afflicts a large population worldwide. Early diagnosis of cognitive impairment can allow for preventative measures to be taken to slow down or prevent the progression to dementia. In this study, we devise an interpretable graph convolutional neural network approach, GCNEnsemble, using both non-clinical variables such as MRI preprocessed features including cortical thickness and gray matter volumes, and clinical features from a community-dwelling Southeast Asian population in Singapore aged between 30 and 95 years from the Biomarker and Cognition study (BIOCIS), to classify participants into cognitively normal, subjective cognitive decline, and mild cognitive impairment. We further conducted ablation studies and varied the quantities of labeled data to understand the contribution of the non-clinical features and the applicability of GCNEnsemble in low to high labeled data availability scenarios. GCNEnsemble was able to attain the highest accuracy and Matthew's correlation coefficient compared to existing state-of-the-art methods. Feature interpretability via Integrated Gradients identified features such as visual cognitive assessment test (VCAT), systolic and diastolic blood pressure, and cerebrospinal fluid volume as key features for the classification, with VCAT having the highest feature importance. There was higher median cerebrospinal fluid volume, right frontal pole thickness, left pallidum volume, and right hippocampal fissure volume but lower VCAT for the mild cognitive impairment group than the two other groups. In conclusion, GCNEnsemble can be used as a semi-supervised interpretable classification tool for cognitive syndrome in a Southeast Asian population.Ministry of Education (MOE)National Medical Research Council (NMRC)Submitted/Accepted versionThis study received funding support from the Strategic Academic Initiative grant from the Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore, National Medical Research Council, Singapore under its Clinician Scientist Award (MOHsingle bondCSAINV18nov-0007), Ministry of Education Start-up Grant and Ministry of Education Academic Research Fund Tier 1 (RT02/21)
A silicon analogue of a fused bicyclic borirene derivative
No full textThe replacement of all carbon atoms in aromatic rings with main-group elements to afford inorganic ring systems is highly desirable due to their distinct aromatic character. However, fused polycyclic main-group element rings are rare and the feasibility of aromaticity in such compounds has yet to be explored. To explore aromaticity in fused polycyclic main-group element rings, a stable di-silicon analogue of fused bicyclic borirene, namely bicyclo[1.1.0]-2,4-diborylenyldisil-1(3)-ene 5 was synthesized from an N-phosphinoamidinato chlorosilylene 3. Compound 5 consists of a bridgehead Si=Si double bond bonded with two bridging borons resulting in an unsaturated fused bicyclic skeleton. The bridgehead Si=Si sand p-electrons and bridging Si–B s-electrons are stabilized by both s- and p-aromatic delocalization on the Si2B2 fused bicyclic ring.Agency for Science, Technology and Research (A*STAR)Ministry of Education (MOE)Published versionThis work was supported by A*STAR MTC Individual Research Grants (M21K2c0117) and the Ministry of Education Singapore AcRF Tier 1 (RG72/21). M.-D. Su acknowledges the Ministry of Science and Technology of Taiwan for the financial support
"Do you not think I am a doctor collecting medicines?”: Henry Nicholas Ridley's Medical Botany in the age of Malayan tropical medicine, 1888-1912
No full textInvestigating Ridley’s medical botany, this paper illustrates that the Director of the Singapore Botanic Gardens, Henry Nicholas Ridley (active 1888–1912), turned towards rather than away from local flora medica in a period when colonial doctors fervently practiced an increasingly biologised British tropical medicine. Ridley along with a select few British physicians and anthropologists, as part of networks of knowledge, therefore pioneered modes of thought on Malaya’s plant-based pharmakon. Though these approaches were largely rooted in contemporary European scientific medicine that understood local substances on a chemical basis and judiciously appropriated only those deemed medicinally potent, they nonetheless breathed an air of legitimacy into “native” medical knowledge systems that most British doctors derided for their magico-abstract nature. This paper therefore contends that Ridley was a dualistic figure who selectively promoted and downplayed local pharmakon in various contexts. His work will be shown to represent a sustained British imperial interest in drugs from the colonies, counterposing the pre-existing historical literature which tends to assume the incompatibility of “civilising” European medical discourses and practices with local therapeutic beliefs from the end of the nineteenth century onwards.Bachelor's degre
Effective fake news detection
No full textThe increasing numbers of fake news on online platforms poses significant societal
challenges, influencing public opinion and undermines confidence in reliable news
sources. This report explores the development of models for detecting fake news,
leveraging text and link analysis. Existing literature demonstrates the advantages of these
approaches, but challenges such as dataset quality, computational efficiency, and
resilience to adversarial attacks persist. By analyzing different datasets and evaluating
models like BERT, RoBERTa and Long Short-Term Memory (LSTM) networks, this
project aims to improve detection accuracy and propose scalable solutions. Preliminary
results indicate that balanced datasets and robust features significantly enhance model
performance, underscoring the potential of this approach to combat misinformation
effectively.Bachelor's degre