Hi! I am an Assistant Professor at the School of Mathematics, University of Birmingham. I am also a visiting scholar at the Cambridge Image Analysis Group, University of Cambridge. My research interests include large-scale optimization, statistical learning theory, explainable AI, with applications in many areas of data science, such as computational and medical imaging. Most recently, my research has been focusing on the theoretical foundations of non-convex optimization in computational imaging with deep-learning image priors, and efficient deep unrolling networks for medical imaging. Meanwhile, I am also diving into the research of large-scale optimization and machine learning for computational social science and financial applications.
In 2019, I completed my Ph.D in the University of Edinburgh under the supervision of Prof. Mike Davies and the project was fully funded by EU H2020 project MacSeNet Innovative Training Network. Prior to that, I had 4 wonderful years in Sichuan University, China as a undergrad majoring in Communication Engineering from 2010 to 2014.
Openings: If you wish to join my team as PhD student (full scholarship available), please email me with your CV and transcripts 🙂
Email: j.tang.2[AT]bham.ac.uk, jt814[AT]cam.ac.uk, JesuslovesBilly [AT] gmail.com
Recent works/preprints:
- Ziruo Cai, Junqi Tang, Subhadip Mukherjee, Jinglai Li, Carola-Bibiane Schönlieb, Xiaoqun Zhang. NF-ULA: Langevin Monte Carlo with Normalizing Flow Prior for Imaging Inverse Problems. Technical Report, 2023 [Preprint]
- Hong Ye Tan, Subhadip Mukherjee, Junqi Tang, Carola-Bibiane Schönlieb. Provably Convergent Plug-and-Play Quasi-Newton Methods. Technical Report, 2023 [Preprint][Invited_talk@AIP2023]
- Derek Driggs, Matthias Ehrhardt, Carola-Bibiane Schönlieb, Junqi Tang*. Practical Acceleration of the Condat-Vu Algorithm. Technical report, 2023 [Preprint][link]
- Antonin Chambolle, Claire Delplancke, Matthias Ehrhardt, Carola-Bibiane Schönlieb, Junqi Tang*. Stochastic Primal-Dual Hybrid Gradient Algorithm with Adaptive Step-Sizes. Technical report, 2023 [Preprint][Presentation@SIAM-MDS]
- Junqi Tang, Subhadip Mukherjee, Carola-Bibiane Schönlieb. Accelerating Deep Unrolling Networks via Dimensionality Reduction. Technical report, 2022 [Preprint][Invited talk@ICSDS]
Publications:
- Hong Ye Tan, Subhadip Mukherjee, Junqi Tang, Carola-Bibiane Schönlieb. Data-Driven Mirror Descent with Input-Convex Neural Networks. SIAM Journal on Mathematics of Data Science, 2023 [PDF]
- Hong Ye Tan, Subhadip Mukherjee, Junqi Tang, Andreas Hauptmann, Carola-Bibiane Schönlieb. Robust Data-Driven Accelerated Mirror Descent. International Conference on Acoustics, Speech and Signal Processing (ICASSP), Invited Paper, 2023 [PDF]
- Bin Qian, Zhenyu Wen, Junqi Tang, Ye Yuan, Albert Zomaya, Rajiv Ranjan. OsmoticGate: Adaptive Edge-based Real-time Video Analytics for the Internet of Things. IEEE Transactions on Computers, 2022 [PDF]
- Derek Driggs*, Junqi Tang*, Jingwei Liang, Mike Davies, Carola-Bibiane Schönlieb. Stochastic Proximal Alternating Minimization for Non-smooth and Non-convex Optimization. SIAM Journal on Imaging Sciences, 2021 [PDF][Code]
- Julian Tachella, Junqi Tang, Mike Davies. The Neural Tangent Link Between CNN Denoisers and Non-local Filters. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR, oral), 2021 [PDF][Code]
- Junqi Tang, Karen Egiazarian, Mohammad Golbabaee, Mike Davies. The Practicality of Stochastic Optimization in Imaging Inverse Problems. IEEE Transactions on Computational Imaging, 2020 [PDF]
- Junqi Tang, Karen Egiazarian, Mike Davies. The Limitation and Practical Acceleration of Stochastic Gradient Algorithms in Inverse Problems. International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2019 [PDF] [slides]
- Junqi Tang, Mohammad Golbabaee, Francis Bach, Mike Davies. Rest-Katyusha: Exploiting the Solution’s Structure via Scheduled Restart Schemes. Advances in Neural Information Processing Systems (NeurIPS), 2018.[PDF] [Poster]
- Junqi Tang, Mohammad Golbabaee, Mike Davies. Exploiting the Structure via Sketched Gradient Algorithms. IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2017.[PDF]
- Junqi Tang, Mohammad Golbabaee, Mike Davies. Gradient Projection Iterative Sketch for Large-Scale Constrained Least-Squares. International Conference on Machine Learning (ICML), 2017.[PDF] [slides]
Thesis:
- Randomized Structure-Adaptive Optimization. PhD thesis, 2019 [link]
- The Non-Uniform Fast Fourier Transform in Computed Tomography. MSc thesis, 2015 [link]
***********************************************************************************
Reviewer of:
ICML 2022, NeurIPS 2021, ICML 2021, CVPR 2021, ICLR 2021, NeurIPS 2020, NeurIPS 2019
IMA Journal on Numerical Analysis,
Numerical Algorithms,
Mathematics of Operations Research,
IEEE Transactions on Neural Networks and Learning Systems
***********************************************************************************
JT 