Journal Articles
High-Order Synchrosqueezed Chirplet Transforms for Multicomponent Signal Analysis
Yi-Ju Yen, De-Yan Lu, “Sing-Yuan Yeh”, Jian-Jiun Ding, and Chun-Yen Shen
Applied and Computational Harmonic Analysis, 82, 101839 (2026)
A higher-order synchrosqueezed chirplet transform improves the analysis of multicomponent signals whose instantaneous frequencies cross, especially under strong chirp modulation.
#signal processing · #time-frequency analysis · #synchrosqueezing · #chirplet transforms
Landmark Alternating Diffusion
“Sing-Yuan Yeh”, Hau-Tieng Wu, Ronen Talmon, and Mao-Pei Tsui
SIAM Journal on Mathematics of Data Science, 7(2), 621–642 (2025)
Landmark alternating diffusion makes multimodal sensor fusion more computationally efficient while retaining its manifold-based structure; the paper includes theory and a sleep-stage annotation application.
#manifold learning · #multimodal data · #sensor fusion · #diffusion maps · #landmark methods
Conference Papers
Sample Complexity of Kernel-Based Q-Learning
“Sing-Yuan Yeh”, Fu-Chieh Chang, Chang-Wei Yueh, Pei-Yuan Wu, Alberto Bernacchia, and Sattar Vakili
Proceedings of the 26th International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 206 (2023)
We analyze kernel-based Q-learning with a generative model and give sample-complexity guarantees for large, potentially infinite state-action spaces.
#reinforcement learning · #kernel methods · #Q-learning · #sample complexity
Preprints
Accelerate Vector Diffusion Maps by Landmarks
“Sing-Yuan Yeh”, Yi-An Wu, Hau-Tieng Wu, and Mao-Pei Tsui
arXiv:2603.21247 (2026)
Landmark-accelerated vector diffusion maps use a two-stage normalization for nonuniform sampling and converge to the connection Laplacian under a manifold model.
#manifold learning · #vector diffusion maps · #connection Laplacian · #landmark methods
Uncertainty of Network Topology with Applications to Out-of-Distribution Detection
“Sing-Yuan Yeh” and Chun-Hao Yang
arXiv:2511.18813; submitted for journal publication (2025)
Predictive topological uncertainty summarizes how a Bayesian neural network interacts with its inputs and supports a significance test for out-of-distribution detection.
#topological data analysis · #Bayesian neural networks · #out-of-distribution detection · #uncertainty quantification · #persistent homology