A5058964100- Advanced Neural Network Applications
- Energy Efficient Wireless Sensor Networks
- Adversarial Robustness in Machine Learning
- Wireless Networks and Protocols
- Domain Adaptation and Few-Shot Learning
- Anomaly Detection Techniques and Applications
- Artificial Immune Systems Applications
- Network Security and Intrusion Detection
- Network Traffic and Congestion Control
- Advanced Memory and Neural Computing
- Cholangiocarcinoma and Gallbladder Cancer Studies
- Distributed Sensor Networks and Detection Algorithms
- Software-Defined Networks and 5G
- Embedded Systems Design Techniques
- Image Processing Techniques and Applications
- Indoor and Outdoor Localization Technologies
- Extracellular vesicles in disease
- IoT and Edge/Fog Computing
- RNA modifications and cancer
- Machine Learning and Data Classification
- Fibroblast Growth Factor Research
- Network Time Synchronization Technologies
- Modular Robots and Swarm Intelligence
- MicroRNA in disease regulation
- Digital Imaging for Blood Diseases
Qilu Hospital of Shandong University
2023
Huawei Technologies (China)
2021
Xi'an Jiaotong University
2011-2016
Huawei Technologies (United Kingdom)
2016
Background & Aims: Cholangiocarcinoma (CCA) is a highly lethal malignancy originating from the biliary ducts. Current CCA diagnostic and prognostic assessments cannot satisfy clinical requirement. Bile detection rarely performed, herein, we aim to estimate significance of bile liquid biopsy by assessing exosomal concentrations components. Approach Results: Exosomes in sera CCA, pancreatic cancer, common duct stone were identified quantified transmission electronmicroscopy, nanoparticle...
Abstract and Aim: Cholangiocarcinoma (CCA) is a highly aggressive lethal cancer that originates from the biliary epithelium. Systemic treatment options for CCA are currently limited, first targeted drug of CCA, pemigatinib, emerged in 2020 by inhibiting FGFR2 phosphorylation. However, regulatory mechanism phosphorylation not fully elucidated. Approach Results: Here we screened FGFR2-interacting proteins showed protein tyrosine phosphatase (PTP) N9 interacts with negatively regulates...
Delay is an important metric to understand and improve system performance. While existing approaches focus on aggregate delay statistics in pre-programmed granularity, providing only statistical results such as averages deviations, those fail provide fine-grained measurement at a flexible level thus may miss characteristics. For example, anomalies, which are critical performance indicators, not be captured by coarse grained approaches. In this work, we propose approach based new structure...
Delay is an important metric to understand and improve system performance. While existing approaches focus on aggregated delay statistics in pre-programmed granularity provide results such as average deviation, those may not fine-grained measurement thus miss characteristics. For example, anomaly, which a critical performance indicator, be captured by coarse-grained approaches. We propose new structure design called order preserving aggregator (OPA). Based OPA, we can efficiently encode...
Recent developments in Neural Architecture Search (NAS) resort to training the supernet of a predefined search space with weight sharing speed up architecture evaluation. These include random schemes, as well various schemes based on optimization or reinforcement learning, particular policy gradient, that aim optimize parametric distribution and shared model weights simultaneously. In this paper, we focus efficiently exploring important region neural learning. We propose Deep Deterministic...
Neural architecture search automates neural network design and has achieved state-of-the-art results in many deep learning applications. While recent literature focused on designing networks to maximize accuracy, little work been conducted understand the compatibility of spaces varying hardware. In this paper, we analyze blocks used build Once-for-All (MobileNetV3), ProxylessNAS ResNet families, order their predictive power inference latency various devices, including Huawei Kirin 9000 NPU,...
Debugging wireless sensor networks (WSNs) is notoriously difficult, due to the resource constraints on sensors and distributed running of debugged programs. Many bugs only manifest themselves during actual operation a network, thus requiring runtime debugging program. A WSN debugger has meet two important design criteria, namely saving energy preserving responsiveness normal system/network events debugging. In this paper, we propose Stethoscope, sustainable for WSNs. devise new technique...
Despite the empirical success of neural architecture search (NAS) in deep learning applications, optimality, reproducibility and cost NAS schemes remain hard to assess. In this paper, we propose Generative Adversarial (GA-NAS) with theoretically provable convergence guarantees, promoting stability search. Inspired by importance sampling, GA-NAS iteratively fits a generator previously discovered top architectures, thus increasingly focusing on important parts large space. Furthermore, an...
Neural architecture search (NAS) has achieved remarkable results in deep neural network design. Differentiable converts the over discrete architectures into a hyperparameter optimization problem which can be solved by gradient descent. However, questions have been raised regarding effectiveness and generalizability of methods for solving non-convex problems. In this paper, we propose L2NAS, learns to intelligently optimize update hyperparameters via an actor based on distribution...
The application of Wireless Sensor Networks (WSNs) often falls into unexpected poor performance conditions due to many factors such as complex network interactions, software bugs and incorrect configurations. Diagnosing a is challenging since it difficult obtain information from the including (1) non-deterministic interactions among motes, (2) difficulties in reconstructing status each individual mote, (3) unavailability real environment information. To address these problems, we propose...
Wireless sensor networks (WSNs) have been extensively applied in many important fields with larger scale and more complex structure. The applications of WSNs are regarded as a sustainable solution to provide ongoing efficient monitoring services the real world. When such an application faces poor performance or unexpected condition, administrator needs deploy diagnosis system diagnose task. One possible way is transform some original motes by using reprogramming technique. However, challenge...
Event detection is a typical application of wireless sensor networks. The existing approaches event usually employ certain models that are constructed with prior domain knowledge. resulting processes appear to be cost-inefficient, which require either intensive data exchanges among neighboring nodes or caching large columns history data. In this paper, we focus on the issue locating in network without locations. This involves two tasks, namely detecting an and identifying area where occurs....
How to reasonably allocate and schedule resources of wireless sensor system maximum its potential capability has been an important area for research. In this paper, we focus on the Optimal Sampling Frequency Assignment (OSFA) in a real-time networks (RTWSN). An appropriate OSFA should both guarantee good quality service efficiently utilize limited network as well. We propose distributed optimization algorithm, called Lazy Schema (LySa), obtain optimal sampling rates source nodes with low...
Despite the empirical success of neural architecture search (NAS) in deep learning applications, optimality, reproducibility and cost NAS schemes remain hard to assess. In this paper, we propose Generative Adversarial (GA-NAS) with theoretically provable convergence guarantees, promoting stability search. Inspired by importance sampling, GA-NAS iteratively fits a generator previously discovered top architectures, thus increasingly focusing on important parts large space. Furthermore, an...