Chunhui Chen

Chunhui Chen

Ph.D. candidate at Tsinghua University

Publications

A numerical algorithm with linear complexity for Multi-marginal Optimal Transport with L1 Cost

Published in Submitted to CSIAM Transactions on Applied Mathmatics, 2024

Numerically solving multi-marginal optimal transport (MMOT) problems is computationally prohibitive, even for moderate-scale instances involving l larger than 4 marginals with support sizes of N larger than 1000.

Recommended citation: C. Chen, J. Chen, B. Luo, S. Jin and H. Wu. " A numerical algorithm with linear complexity for Multi-marginal Optimal Transport with L1 Cost." arXiv preprint arXiv:2405.19246 (2024). https://arxiv.org/pdf/2405.19246

Computation and Critical Transitions of Rate-Distortion-Perception Functions With Wasserstein Barycenter

Published in Submitted to IEEE Transactions on Information Theory, 2024

The information rate-distortion-perception (RDP) function characterizes the three-way trade-off between description rate, average distortion, and perceptual quality measured by discrepancy between probability distributions.

Recommended citation: C. Chen, X. Niu, and W. Hao. " Computation and Critical Transitions of Rate-Distortion-Perception Functions With Wasserstein Barycenter." arXiv preprint arXiv:2404.04681 (2024). https://arxiv.org/pdf/2404.04681

Computation of Rate-Distortion-Perception Functions With Wasserstein Barycenter

Published in IEEE International Symposium on Information Theory (ISIT), 2023., 2023

In this paper, we show that the information RDP function can be transformed into a Wasserstein Barycenter problem. The nonstrictly convexity brought by the perceptual constraint can be regularized by an entropy regularization term.

Recommended citation: C. Chen, X. Niu, and W. Ye. " Computation of Rate-Distortion-Perception Functions With Wasserstein Barycenter." 2023 IEEE International Symposium on Information Theory (ISIT). https://ieeexplore.ieee.org/document/10206982

TC-MIMONet: A Learning-based Transceiver for MIMO Systems with Temporal Correlations

Published in 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring), 2021

In this work, we propose a novel learning-based MIMO transceiver, namely, the TC-MIMONet, which extends the conventional memoryless AE-based transceivers by customizing two neural network components with memory.

Recommended citation: C. Chen, Z. Wang, and Y. Mao. " TC-MIMONet: A Learning-Based Transceiver for MIMO Systems with Temporal Correlations." 2021 IEEE 93rd Vehicular Technology Conference (VTC2021-Spring). https://ieeexplore.ieee.org/abstract/document/9448981