Source author record

Dan Sun

Dan Sun appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Machine Learning Artificial Intelligence cond-mat.str-el Applications cond-mat.supr-con Distributed, Parallel, and Cluster Computing physics.app-ph

Catalog footprint

What is connected

7works

7topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2023arXiv

Anonymous Pattern Molecular Fingerprint and its Applications on Property Identification

Molecular fingerprints are significant cheminformatics tools to map molecules into vectorial space according to their characteristics in diverse functional groups, atom sequences, and other topological structures. In this paper, we set out to investigate a novel molecular fingerprint \emph{Anonymous-FP} that possesses abundant perception about the underlying interactions shaped in small, medium, and large molecular scale links. In detail, the possible inherent atom chains are sampled from each molecule and are extended in a certain anonymous pattern. After that, the molecular fingerprint \emph{Anonymous-FP} is encoded in virtue of the Natural Language Processing technique \emph{PV-DBOW}. \emph{Anonymous-FP} is studied on molecular property identification and has shown valuable advantages such as rich information content, high experimental performance, and full structural significance. During the experimental verification, the scale of the atom chain or its anonymous manner matters significantly to the overall representation ability of \emph{Anonymous-FP}. Generally, the typical scale $r = 8$ enhances the performance on a series of real-world molecules, and specifically, the accuracy could level up to above $93\%$ on all NCI datasets.

preprint2022arXiv

A Graph Data Augmentation Strategy with Entropy Preservation

The Graph Convolutional Networks (GCN) proposed by Kipf and Welling is an effective model for semi-supervised learning, but faces the obstacle of over-smoothing, which will weaken the representation ability of GCN. Recently some works are proposed to tackle above limitation by randomly perturbing graph topology or feature matrix to generate data augmentations as input for training. However, these operations inevitably do damage to the integrity of information structures and have to sacrifice the smoothness of feature manifold. In this paper, we first introduce a novel graph entropy definition as a measure to quantitatively evaluate the smoothness of a data manifold and then point out that this graph entropy is controlled by triangle motif-based information structures. Considering the preservation of graph entropy, we propose an effective strategy to generate randomly perturbed training data but maintain both graph topology and graph entropy. Extensive experiments have been conducted on real-world datasets and the results verify the effectiveness of our proposed method in improving semi-supervised node classification accuracy compared with a surge of baselines. Beyond that, our proposed approach could significantly enhance the robustness of training process for GCN.

preprint2022arXiv

High-temperature superconductivity in hydrides: experimental evidence and details

Since the discovery of superconductivity at 200 K in H3S [1] similar or higher transition temperatures, Tcs, have been reported for various hydrogen-rich compounds under ultra-high pressures [2]. Superconductivity was experimentally proved by different methods, including electrical resistance, magnetic susceptibility, optical infrared, and nuclear resonant scattering measurements. The crystal structures of superconducting phases were determined by X-ray diffraction. Numerous electrical transport measurements demonstrate the typical behaviour of a conventional phonon-mediated superconductor: zero resistance below Tc, the shift of Tc to lower temperatures under external magnetic fields, and pronounced isotope effect. Remarkably, the results are in good agreement with the theoretical predictions, which describe superconductivity in hydrides within the framework of the conventional BCS theory. However, despite this acknowledgment, experimental evidence for the superconducting state in these compounds has recently been treated with criticism [3, 4], which apparently stems from misunderstanding and misinterpretation of complicated experiments performed under very high pressures. Here, we describe in greater detail the experiments revealing high-temperature superconductivity in hydrides under high pressures. We show that the arguments against superconductivity [3, 4] can be either refuted or explained. The experiments on the high-temperature superconductivity in hydrides clearly contradict the theory of hole superconductivity [4] and eliminate it [3].

preprint2021arXiv

Composite Pressure Cell for Pulsed Magnets

Extreme pressures and high magnetic fields can affect materials in profound and fascinating ways. However, large pressures and fields are often mutually incompatible; the rapidly changing fields provided by pulsed magnets induce eddy currents in the metallic components used in conventional pressure cells, causing serious heating, forces and vibration. Here we report a diamond-anvil-cell made mainly out of insulating composites that minimizes inductive heating while retaining sufficient strength to apply pressures of up to 9 GPa. Any residual metallic components are made of low-conductivity metals and patterned to reduce eddy currents. The simple design enables rapid sample or pressure changes, desired by pulsed-magnetic-field-facility users. The pressure cell has been used in pulsed magnetic fields of up to 65 T with no noticeable heating at cryogenic temperatures. Several measurement techniques are possible inside the cell at temperatures as low as 500 mK.

preprint2021arXiv

Graph Classification Based on Skeleton and Component Features

Most existing popular methods for learning graph embedding only consider fixed-order global structural features and lack structures hierarchical representation. To address this weakness, we propose a novel graph embedding algorithm named GraphCSC that realizes classification based on skeleton information using fixed-order structures learned in anonymous random walks manner, and component information using different size subgraphs. Two graphs are similar if their skeletons and components are both similar, thus in our model, we integrate both of them together into embeddings as graph homogeneity characterization. We demonstrate our model on different datasets in comparison with a comprehensive list of up-to-date state-of-the-art baselines, and experiments show that our work is superior in real-world graph classification tasks.

preprint2020arXiv

Serverless inferencing on Kubernetes

Organisations are increasingly putting machine learning models into production at scale. The increasing popularity of serverless scale-to-zero paradigms presents an opportunity for deploying machine learning models to help mitigate infrastructure costs when many models may not be in continuous use. We will discuss the KFServing project which builds on the KNative serverless paradigm to provide a serverless machine learning inference solution that allows a consistent and simple interface for data scientists to deploy their models. We will show how it solves the challenges of autoscaling GPU based inference and discuss some of the lessons learnt from using it in production.

preprint2014arXiv

Pressure study of nematicity and quantum criticality in Sr$_3$Ru$_2$O$_7$ for an in-plane field

We study the relationship between the nematic phases of Sr$_3$Ru$_2$O$_7$ and quantum criticality. At ambient pressure, one nematic phase is associated with a metamagnetic quantum critical end point (QCEP) when the applied magnetic field is near the \textit{c}-axis. We show, however, that this metamagnetic transition does not produce the same nematic signatures when the QCEP is reached by hydrostatic pressure with the field applied in the \textit{ab}-plane. Moreover, a second nematic phase, that is seen for field applied in the \textit{ab}-plane close to, but not right at, a second metamagnetic anomaly, persists with minimal change to the highest applied pressure, 16.55 kbar. Taken together our results suggest that metamagnetic quantum criticality may not be necessary for the formation of a nematic phase in Sr$_3$Ru$_2$O$_7$.