Variational quantum evolution equation solver

Consider the second-order homogeneous evolution equation defined on the set \(\Omega \times J\), where \(\Omega \subset \mathbb R^d\) denotes a d-dimensional bounded spatial domain and \(J = [0,T]\), where \(T>0\) denotes a bounded temporal domain, as $$\begin{aligned} \frac{\partial u(\vec{x},t)}{\partial t}&= D \nabla ^{2} u(\vec{x},t) + f(\vec{x},t), \qquad&\text{ in } \Omega \times J \end{aligned}$$ (1) $$\begin{aligned}… Continue reading Variational quantum evolution equation solver

Three MIT faculty members named 2022 Simons Investigators | MIT News

Physics professors Tracy Slatyer and Jesse Thaler and mathematics professor Wei Zhang were among 16 researchers selected to receive 2022 Simons Investigator awards in mathematics, physics, astrophysics, and computer science. This program supports “outstanding theoretical scientists in their most productive years, when they are establishing creative new research directions, providing leadership to the field, and effectively mentoring… Continue reading Three MIT faculty members named 2022 Simons Investigators | MIT News

Researchers develop a model of motor learning that is able to simulate the results of experiments in humans — ScienceDaily

Researchers from the University of Tsukuba have developed a mathematical model of motor learning that reflects the motor learning process in the human brain. Their findings suggest that motor exploration — that is, increased variability in movements — is important when learning a new task. These results may lead to improved motor rehabilitation in patients… Continue reading Researchers develop a model of motor learning that is able to simulate the results of experiments in humans — ScienceDaily

Label-free flow cytometry of rare circulating tumor cell clusters in whole blood

In vitro microfluidic cell sorting has improved the isolation of CTCs and CTCCs from whole blood samples. At the forefront of cell sorting microfluidic chips are the DLD chip and NISA-XL chip13,14. The DLD chip capitalized on the size and asymmetry of CTCCs to isolate them from whole blood samples. This method showed a sorting… Continue reading Label-free flow cytometry of rare circulating tumor cell clusters in whole blood

Graph convolutional networks fusing motif-structure information

The essence of convolutional operations in graph convolutional networks is to achieve feature learning through message aggregation of each node and its first-order neighbors. As shown in Fig. 3, when the central node aggregates neighbor features, its feature information comes not only from itself but also from the features of neighboring nodes. Node feature update combines… Continue reading Graph convolutional networks fusing motif-structure information

Model architecture can transform catastrophic forgetting into positive transfer

Learning the ones and twos addition facts: the problem proposed by McCloskey and Cohen In the problem proposed by McCloskey and Cohen, a neural network is trained using two tasks: the “ones addition facts” (all the additions of 1 with another digit, \(1+1=2\) through \(9+1=10\) and \(1+1=2\) through \(1+9=10\)) and the “twos addition facts” (\(1+2=3\)… Continue reading Model architecture can transform catastrophic forgetting into positive transfer

Preparing for the next pandemic via transfer learning from existing diseases with hierarchical multi-modal BERT: a study on COVID-19 outcome prediction

We begin this section with a description of the available data sources and the cohort selection process. After that, the problem statement is described followed by a description of the model architecture. Data sources Our study is based on de-identified EHR data of all patients treated at Stanford Hospital, between January 1, 2015 and March… Continue reading Preparing for the next pandemic via transfer learning from existing diseases with hierarchical multi-modal BERT: a study on COVID-19 outcome prediction

Kolmogorov compression complexity may differentiate different schools of Orthodox iconography

Entropy and nine fractal parameters were employed in the exploration of the 1200 byzantine icon images, as well as artificially generated fractal surface (FS) images, to validate our methodology. Generating test images To validate the proposed approach, we generated 5 × 11 sets of generated images with fractal surfaces using an Inverse FFT algorithm and fractal dimensions… Continue reading Kolmogorov compression complexity may differentiate different schools of Orthodox iconography

Inferring the spread of COVID-19: the role of time-varying reporting rate in epidemiological modelling

S(E)IR epidemiological model To investigate the spread of COVID-19 through a population, we use the traditional SIR and SEIR epidemiological models, augmenting them with two temporal variables. One of these variables accounts for the rate of transmission, while the other variable quantifies the rate at which positive cases are reported. These time-varying variables account for… Continue reading Inferring the spread of COVID-19: the role of time-varying reporting rate in epidemiological modelling

GIS-based precise predictive model of mountain beacon sites in Wenzhou, China

Study area The Wenzhou (温州) area was chosen as the study area, including the inner city of Wenzhou, as well as the counties and villages under its jurisdiction, with a total area of approximately 12,110 km222. Wenzhou is located in East China, southeast of Zhejiang Province, lower reaches of the Ou River, bordering the East… Continue reading GIS-based precise predictive model of mountain beacon sites in Wenzhou, China