A5021421021- Machine Learning in Materials Science
- Metal-Organic Frameworks: Synthesis and Applications
- X-ray Diffraction in Crystallography
- Reinforcement Learning in Robotics
- Catalytic Processes in Materials Science
- Single-cell and spatial transcriptomics
- Catalysis and Oxidation Reactions
- Supply Chain and Inventory Management
- Bioinformatics and Genomic Networks
- Thermochemical Biomass Conversion Processes
- Traffic control and management
- Innovation Diffusion and Forecasting
- Health, Environment, Cognitive Aging
- Subtitles and Audiovisual Media
- Combustion and Detonation Processes
- Advanced Control Systems Optimization
- Fault Detection and Control Systems
- Corrosion Behavior and Inhibition
- Hydrogen Storage and Materials
- Advanced Combustion Engine Technologies
- Ammonia Synthesis and Nitrogen Reduction
- Covalent Organic Framework Applications
- Catalysts for Methane Reforming
- Natural Language Processing Techniques
- Educational Systems and Policies
National University of Singapore
2022-2025
Spatial omics data are clustered to define both cell types and tissue domains. We present Building Aggregates with a Neighborhood Kernel Yardstick (BANKSY), an algorithm that unifies these two spatial clustering problems by embedding cells in product space of their own the local neighborhood transcriptome, representing state microenvironment, respectively. BANKSY's feature augmentation strategy improved performance on tasks when tested diverse RNA (imaging, sequencing) protein (imaging)...
The search for novel catalytic processes is essential to combat climate change by tapping into sustainable hydrogen technologies, which rely on low-carbon H2 production and efficient storage global transportation. Ammonia, can function as a carrier later dehydrogenation in COx-free production, recognized one of the key solutions. utilization nonthermal plasma catalyze reactions plausible means replacing CO2-polluting fossil-dependent thermal processes. In this perspective, we summarize...
Digital discovery of functional materials, such as metal–organic frameworks (MOFs), entails accurate and data-efficient approaches to navigate complex chemical structural space. Based on an innovative deep learning approach, namely, Kolmogorov–Arnold Networks (KANs), we introduce MOF-KAN, a state-of-the-art architecture the first application KANs digital MOFs. Through meticulous fine-tuning network architecture, demonstrate that MOF-KAN outperforms standard multilayer perceptrons (MLPs) in...
Poor water stability of many metal–organic frameworks (MOFs) remains a persistent bottleneck toward their practical applications. Recently, hemilabile STAMs experimentally demonstrate much stronger stability, as well improved adsorption performance under humid conditions, than the compositionally similar HKUST-1. Yet, fundamental properties remain largely unexplored. Herein, we apply density-functional theory (DFT) calculations and machine-learned potentials (MLP) based molecular dynamics...
Metal-organic frameworks (MOFs) are versatile materials for a wide variety of potential applications. Tunable thermal expansion properties promote the application MOFs in thermally sensitive composite materials; however, they currently available only handful structures. Herein, we report first data set 33,131 diverse generated from molecular simulations and subsequently develop machine learning (ML) models to (1) classify different behaviors (2) predict volumetric coefficients (α
Machine-learned potentials (MLPs) have transformed the field of molecular simulations by scaling "quantum-accurate" to linear time complexity. While they provide more accurate reproduction physical properties as compared empirical force fields, it is still computationally costly generate their training data sets from ab initio calculations. Despite emergence foundational or general MLPs for organic molecules and dense materials, unexplored if one MLP can be effectively developed a wide...
Abstract The application of reinforcement learning (RL) in process control has garnered increasing research attention. However, much the current literature is focused on training and deploying a single RL agent. multi‐agent (MARL) not been fully explored control. This work aims to: (i) develop unique agent configuration that suitable MARL system for multiloop control, (ii) demonstrate efficacy systems controlling even exhibit strong interactions, (iii) conduct comparative study performance...
Metal-organic frameworks (MOFs) represent a distinctive class of nanoporous materials with considerable potential across wide range applications. Recently, handful MOFs has been explored for the storage environmentally hazardous fluorinated gases (Keasler et al.
Reinforcement Learning (RL) provide a model-free method of controlling processes. Recent advancements in Deep have developed RL agents that can potentially control multivariate Multi-Agent (MARL) is subfield where multiple are trained shared environment. In this study, MARL system comprising Twin-Delayed Deterministic Policy Gradient (TD3) to multiloop CSTR process. The achieves stable closed-loop response and good disturbance rejection.
Machine-learned Potentials (MLPs) have transformed the field of molecular simulations by scaling `quantum-accurate' potentials to linear time complexity. Yet, while they provide a more accurate reproduction structural properties as compared empirical force fields, it is still computationally costly generate their training dataset from \textit{ab initio} calculations. However, in current literature, one MLP model always specifically developed and employed for specific system, unexplored if...
Metal–organic frameworks (MOFs) are versatile nanoporous materials for a wide variety of important applications. Recently, handful MOFs have been explored the storage toxic fluorinated gases (Keasler et al. Science, 2023, 381, 1455), yet potential great number such an environmentally sustainable application has not thoroughly investigated. In this work, we apply active learning (AL) to accelerate discovery hypothetical (hMOFs) that can efficiently store specific gas, namely, vinylidene...
On the basis of Markedness Theory, first language transfer phenomena between object sentences in Chinese and Korean process acquisition was researched. During research, a translation test conducted terms four types sentences, i.e. intransitive verb with object, transitive Ba-sentences adjective object. The following two purposes were aimed to be achieved through test. Firstly, understand errors learners China learning sentences. Secondly, track changes Korea Due differences degree amount...