Qiao Ke is an Assistant Professor at Northwestern Polytechnical University, specializing in Deep Learning, Machine Learning, Statistics Learning, Intelligent Software Engineering, and Internet of Things. Qiao holds a Ph.D. in Mathematics from Xi’an Jiao Tong University and has been actively engaged in research, contributing significantly to various areas of computational mathematics and artificial intelligence.
Author Metrics:
Ke, Qiao – Scopus Profile
Orcid Profile
Qiao Ke is affiliated with Northwestern Polytechnical University in Xi’an, China. The Scopus Author Identifier 56465532300 provides valuable metrics regarding their academic contributions.
- Citations: Qiao Ke has received a total of 481 citations across 420 documents, indicating the impact of their research on the academic community.
- Documents: The author has contributed to 16 documents, showcasing a consistent and substantive scholarly output.
- h-index: With an h-index of 8, Qiao Ke has demonstrated a noteworthy level of influence in their field. The h-index is a metric that considers both the number of publications and the number of citations they receive.
These metrics reflect the academic impact and productivity of Qiao Ke, highlighting their contributions to the scholarly landscape. The provided information encourages further exploration into the specific content and context of their publications for a comprehensive understanding of their research achievements.
Education:
Qiao Ke pursued a B.S. in Mathematics from Shaanxi Normal University, an M.S. in Mathematics, and a Ph.D. in Mathematics from Xi’an Jiao Tong University. Additionally, they completed postdoctoral research in the Department of Computer Science at Northwestern Polytechnical University.
Research Focus:
Qiao Ke’s research interests span Deep Learning, Machine Learning, Statistics Learning, Intelligent Software Engineering, and the Internet of Things. Notably, their work includes innovative contributions to neural frameworks for software models, hierarchical search-based code generation, and adaptive disentangled representation learning.
Professional Journey:
Qiao Ke’s professional journey involves serving as an Assistant Professor at the School of Mathematics and Statistics, Northwestern Polytechnical University. They have also actively participated as a reviewer for several reputed journals and conferences, demonstrating their commitment to scholarly peer review.
Publications Top Noted & Contributions:
Qiao Ke has made significant contributions to the field, with publications in respected journals and conferences. Notable works include research on modular neural frameworks for software model connections, deep hierarchical search-based code generation, and adaptive disentangled representation learning.
A research paper titled “RRGcode: Deep hierarchical search-based code generation.” The paper addresses the challenges of retrieval-augmented code generation, where a retrieval model is used to select relevant code snippets from a code corpus to strengthen the generation model. The primary concern is that if the retrieval corpus contains errors or sub-optimal examples, the generation model might replicate these mistakes in the generated code.
To overcome these challenges, the authors propose RRGcode, a deep hierarchical search-based code generation framework. The key components of RRGcode are outlined as follows:
- Retrieval: The framework first retrieves relevant code candidates from a large code corpus. This initial retrieval step aims to gather a set of potential code snippets based on the given query.
- Re-ranking: A re-ranking model is introduced to fine-tune the initial retrieved code rankings. This involves a detailed semantic comparison between the code candidates and the query, ensuring that only the most relevant and accurate candidates are considered. The re-ranking process aims to mitigate the risk of replicating errors from the retrieval corpus.
- Generation: The re-ranked top-K codes, along with the query, serve as input for the code generation model. This final step focuses on generating high-quality and reliable code based on the refined set of code candidates.
The authors claim that RRGcode demonstrates state-of-the-art performance in code generation tasks through extensive experiments. The deep hierarchical search-based approach aims to improve the quality of generated code by addressing the limitations associated with erroneous or sub-optimal code examples present in the retrieval corpus.
1. Title: Spline Interpolation and Deep Neural Networks as Feature Extractors for Signature Verification Purposes
2. Title: Intelligent Internet of Things System for Smart Home Optimal Convection
- Publication Date: June 2021
- Journal: IEEE Transactions on Industrial Informatics
- DOI: 10.1109/tii.2020.3009094
- ISSN: 1551-3203, 1941-0050
3. Title: High-Resolution SAR Image Despeckling Based on Nonlocal Means Filter and Modified AA Model
- Publication Date: November 28, 2020
- Journal: Security and Communication Networks
- DOI: 10.1155/2020/8889317
- ISSN: 1939-0122, 1939-0114
4. Title: Accurate and Fast URL Phishing Detector: A Convolutional Neural Network Approach
5. Title: Adaptive Independent Subspace Analysis of Brain Magnetic Resonance Imaging Data
Research Timeline:
Qiao Ke’s research journey spans from their Bachelor’s degree at Shaanxi Normal University in 2012 to their current role as an Assistant Professor at Northwestern Polytechnical University. Notable milestones include completing a Ph.D., engaging in postdoctoral research, and actively contributing to various research projects, including leadership roles in national and provincial-level foundations.
