Xi Chen (陈熹)
CoRE 735, Busch Campus
Rutgers University
Piscataway, NJ, 08854
I am a Ph.D. candidate advised by Prof. Shirin Jalali at the Electrical and Computer Engineering Department of Rutgers University. My background is in imaging inverse problems, generative models, computational imaging and statistical machine learning.
I am particularly interested in developing denoisers/regularizers (e.g. generative models: VAE, Diffusion models; image implicit/explicit priors: sparsity, DIP, INR; compression code: neural compression etc.), under an optimization framework (e.g. ADMM, Plug-and-Play, SURE, RED etc.), to solve ill-posed inverse problems in 2D/3D imaging systems (e.g. coherent imaging, snapshot compressive imaging etc.), with theoretical guarantees, and understandings of the fundamental limits on the achievable recovery performance from the information theoretic and high-dimensional perspective.
I received my M.Sc. degree in Data Science from Tufts University, MA, worked with Prof. Mike Hughes and Liping Liu. I received my B.Sc. degree in Electrical Engineering from Beijing Institute of Technology, China, worked with Prof. Jie Tian.
Email: firstname.chen15 [at] rutgers [dot] edu
Please feel free to reach out if you want to discuss or collaborate on research!
news
Nov 04, 2024 | Recognized as a Top Reviewer by NeurIPS 2024. |
---|---|
Sep 25, 2024 | Our paper is accepted by NeurIPS 2024. |
Jun 17, 2024 | Glad to receive the SPIE student travel award to present our work on novel approach for coherent imaging system in SPIE Optics + Photonics 2024, San Diego. |
May 20, 2024 | Presented our work on compressive coherent imaging system at Conference on Inverse Problems for Partial Differential Equations. |
May 01, 2024 | Our paper is accepted by ICML 2024. |
Apr 04, 2024 | Presented our work on image acquisition with speckle noise on Columbia Data Science Day. |
Dec 16, 2023 | Presented our work in the NeurIPS Deep learning and Inverse Problems workshop. |
selected publications
- arXivDeCompress: Denoising via Neural CompressionarXiv preprint, 2025
- SampTADeep Memory Unrolled Networks for Solving Imaging Linear Inverse ProblemsInternational Conference on Sampling Theory and Applications, 2025 (Oral)
- ISBIMonte Carlo efficient image reconstruction in coherent imaging with speckle noiseIn IEEE International Symposium on Biomedical Imaging , 2025 (Oral)
- SPIENovel approach to coherent imaging in the presence of speckle noiseIn Unconventional Imaging, Sensing, and Adaptive Optics , 2024 (Oral)
- NeurIPSUntrained Neural Nets for Snapshot Compressive Imaging: Theory and AlgorithmsIn Advances in Neural Information Processing Systems , 2024
- ICMLBagged Deep Image Prior for Recovering Images in the Presence of Speckle NoiseIn International Conference on Machine Learning , 2024
- NeurIPSWMultilook compressive sensing in the presence of speckle noiseIn NeurIPS 2023 Workshop on Deep Learning and Inverse Problems , 2023
- TMLRInterpretable Node Representation with Attribute DecodingTransactions on Machine Learning Research, 2022
- JournalRadiomics analysis of magnetic resonance imaging improves diagnostic performance of lymph node metastasis in patients with cervical cancerRadiotherapy and Oncology, 2019
- SPIERadiomics analysis on T2-MR image to predict lymphovascular space invasion in cervical cancerIn Medical Imaging 2019: Computer-Aided Diagnosis , 2019
- ISMRMRadiomics Analysis of tumor and peri-tumor tissue on T2-Weighted Imaging Improves Diagnostic Performance of Lymph Node Metastasis in Patients with Cervical CancerInternational Society for Magnetic Resonance in Medicine, 2019
- EMBCUnsupervised deep learning features for lung cancer overall survival analysisIn 40th Annual international conference of the IEEE engineering in medicine and biology society (EMBC) , 2018