Kejun Huang
Research Interests
I'm interested in the general area of Machine Learning, Signal Processing, Optimization, and Statistics. Some specific topics include:
Latent variable models with identifiability guarantees, which enables us to discover patterns in an unsupervised fashion, including nonnegative matrix factorization, bounded/independent component analysis, dictionary learning, and tensor decomposition.
Non-convex optimization algorithms that are computationally efficient and (hopefully) with optimality guarantees.
Applications in various areas, including data analytics, natural language processing, and computer vision.
News
Feb. 2023, NSF CAREER Award
Dec. 2022, 2022 IEEE Signal Processing Society Donald G. Fink Overview Paper Award: N. D. Sidiropoulos, L. De Lathauwer, X. Fu, K. Huang, E. E. Papalexakis, and C. Faloutsos, “Tensor Decomposition for Signal Processing and Machine Learning”, IEEE Transactions on Signal Processing, vol. 65, no. 13, pp. 3551-3582, July 1, 2017.
Aug. 2022, SIGBio ACM-BCB Best Student Paper Award to A. Bumin for ‘‘FiT: Fiber-based Tensor Completion for Drug Repurposing’’ by A. Bumin, A. Ritz, D. Slonim, T. Kahveci, and K. Huang (2022).
|