Source author record

Jinjin Zhao

Jinjin Zhao 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 astro-ph.EP Computation and Language Computer Vision cond-mat.mtrl-sci Human-Computer Interaction physics.app-ph

Catalog footprint

What is connected

4works

8topics

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.

preprint2022arXiv

Antiphase boundary in CH$_3$NH$_3$PbI$_3$ repels charge carriers while promotes fast ion migrations

Defects in organic-inorganic hybrid perovskites (OIHPs) greatly influence their optoelectronic properties. Identification and better understanding of defects existing in OIHPs is an essential step towards fabricating high-performance perovskite solar cells. However, direct visualizing the defects is still a challenge for OIHPs due to their sensitivity during electron microscopy characterizations. Here, by using low dose scanning transmission electron microscopy techniques, we observe the common existence of antiphase boundary (APB) in CH$_3$NH$_3$PbI$_3$ (MAPbI$_3$), resolve its atomic structure, and correlate it to the electrical/ionic activities and structural instabilities. Such an APB is caused by the half-unit-cell shift of [PbI$_6$]$_4$-octahedron along the [100]/[010] direction, leading to the transformation from corner-sharing [PbI$_6$]$_4$-octahedron in bulk MAPbI$_3$ into edge-sharing ones at the APB. Based on the identified atomic-scale configuration, we further carry out density functional theory calculations and reveal that the APB in MAPbI$_3$ repels both electrons and holes while serves as a fast ion-migration channel, causing a rapid decomposition into PbI$_2$ that is detrimental to optoelectronic performance. These findings provide valuable insights into the relationships between structures and optoelectronic properties of OIHPs and suggest that controlling the APB is essential for their stability.

preprint2022arXiv

Deep Transfer Learning with Graph Neural Network for Sensor-Based Human Activity Recognition

The sensor-based human activity recognition (HAR) in mobile application scenarios is often confronted with sensor modalities variation and annotated data deficiency. Given this observation, we devised a graph-inspired deep learning approach toward the sensor-based HAR tasks, which was further used to build a deep transfer learning model toward giving a tentative solution for these two challenging problems. Specifically, we present a multi-layer residual structure involved graph convolutional neural network (ResGCNN) toward the sensor-based HAR tasks, namely the HAR-ResGCNN approach. Experimental results on the PAMAP2 and mHealth data sets demonstrate that our ResGCNN is effective at capturing the characteristics of actions with comparable results compared to other sensor-based HAR models (with an average accuracy of 98.18% and 99.07%, respectively). More importantly, the deep transfer learning experiments using the ResGCNN model show excellent transferability and few-shot learning performance. The graph-based framework shows good meta-learning ability and is supposed to be a promising solution in sensor-based HAR tasks.

preprint2020arXiv

Knowledge Graph semantic enhancement of input data for improving AI

Intelligent systems designed using machine learning algorithms require a large number of labeled data. Background knowledge provides complementary, real world factual information that can augment the limited labeled data to train a machine learning algorithm. The term Knowledge Graph (KG) is in vogue as for many practical applications, it is convenient and useful to organize this background knowledge in the form of a graph. Recent academic research and implemented industrial intelligent systems have shown promising performance for machine learning algorithms that combine training data with a knowledge graph. In this article, we discuss the use of relevant KGs to enhance input data for two applications that use machine learning -- recommendation and community detection. The KG improves both accuracy and explainability.

preprint2015arXiv

Photochemical Escape of Oxygen from Early Mars

Photochemical escape is an important process for oxygen escape from present Mars. In this work, a 1-D Monte-Carlo Model is developed to calculate escape rates of energetic oxygen atoms produced from O2+ dissociative recombination reactions (DR) under 1, 3, 10, and 20 times present solar XUV fluxes. We found that although the overall DR rates increase with solar XUV flux almost linearly, oxygen escape rate increases from 1 to 10 times present solar XUV conditions but decreases when increasing solar XUV flux further. Analysis shows that atomic species in the upper thermosphere of early Mars increases more rapidly than O2+ when increasing XUV fluxes. While the latter is the source of energetic O atoms, the former increases the collision probability and thus decreases the escape probability of energetic O. Our results suggest that photochemical escape be a less important escape mechanism than previously thought for the loss of water and/or CO2 from early Mars.