Publications
Preprints/In Preparation
Dirichlet Active Learning. (with Ryan Murray). Submitted to JMLR. [arxiv]
Graph-based Active Learning for Surface Water and Sediment Detection in Multispectral Images. (with Bohan Chen, Jon Schwenk, and Andrea L. Bertozzi). Submitted to IEEE IGARSS 2023.
2023
Cluster-aware Semi-supervised Learning: Relational Knowledge Distillation Provably Learns Clustering. (with Yijun Dong, Qi Lei, and Rachel Ward). Accepted for poster at NeuRIPS 2023. [OpenReview]
Poisson reweighted Laplacian uncertainty sampling for graph-based active learning. (with Jeff Calder). Accepted for publication at SIAM Journal on Mathematics of Data Science. [arxiv]
Model Change Active Learning for Graph-Based Semi-Supervised Learning. (with Andrea L. Bertozzi). Accepted for publication at Springer Journal of Communications on Applied Mathematics and Computation (CAMC). [arxiv]
Batch active learning for multispectral and hyperspectral image segmentation using similarity graphs. (with Bohan Chen, Andrea Bertozzi, and Jon Schwenk). Accepted for publication at Springer Journal of Communications on Applied Mathematics and Computation (CAMC).
Novel Batch Active Learning Approach and Its Application to Synthetic Aperture Radar Datasets. (with James Chapman, Bohan Chen, Zheng Tan, Jeff Calder, and Andrea L. Bertozzi). To appear in Proc. SPIE 12520, Algorithms for Synthetic Aperture Radar Imagery XXX, 12520-13 (2 May 2023).
2022
Efficient and Reliable Overlay Networks for Decentralized Federated Learning. (co-lead author with Yifan Hua, with collaborators Bao Wang, Chen Qian, and Andrea Bertozzi). SIAM Journal on Applied Mathematics}, Volume 82, Number 4, pp. 1558-1586. [journal][arxiv]
Graph-based active learning for semi-supervised classification of SAR data. (lead author, with collaborators Jack Mauro, Jason Setiadi, Xoaquin Baca, Zhan Shi, Jeff Calder, Andrea L. Bertozzi). Proc. SPIE 12095, Algorithms for Synthetic Aperture Radar Imagery XXIX, 120950C (31 May 2022). [proceedings][arxiv]
2021
Posterior Consistency of Semi-Supervised Regression on Graphs. (co-lead author with Hao Li, with collaborators Bamdad Hosseini, Andrew Stuart, and Andrea Bertozzi). Inverse Problems. [journal] [arxiv]
2020
Efficient Graph-Based Active Learning with Probit Likelihood via Gaussian Approximations. (lead author, with collaborators Hao Li and Andrea Bertozzi). ICML 2020 Workshop on Real-World Experiment Design and Active Learning. [arxiv]
2017
Pre-processing and Classification of Hyperspectral Imagery via Selective Inpainting. (with Victoria Chayes, Rasika Bhalerao, Jerry Luo, Wei Shu, Andrea Bertozzi, Wenzhi Liao, and Stanley Osher). IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6195-6199. [ieee]
Dissertation
Active Learning and Uncertainty In Graph-Based Semi-Supervised Learning. UCLA Dissertation for Mathematics Ph.D. [view]