IDEA 团队

首页 > IDEA 团队 > 郑立中
郑立中

郑立中

IDEA研究院院长、首席科学家

教育背景

清华大学电子工程系本硕
加州大学伯克利分校电子工程与计算机科学系博士

研究领域

信息论、无线通信、机器学习和数据科学等领域

电子邮件

zhenglizhong@idea.edu.cn

个人简介

郑立中教授任粤港澳大湾区数字经济研究院院长、首席科学家,同时担任麻省理工学院电气工程与计算机科学系Andrew (1956) and Erna Viterbi教授、香港科技大学访问教授、清华大学杰出访问教授、IEEE Fellow,现任《IEEE信息论汇刊》主编,长期从事信息论、无线通信、机器学习和数据科学等领域的研究,在相关领域顶级学术会议和期刊上发表百余篇论文。曾先后担任腾讯访问科学家、富士康杰出访问科学家、华为/高通/夏普实验室等咨询顾问,目前正积极推动人工智能与相关工程领域的结合,打造跨学科人工智能方法的应用范式,创新科研机制,推动产业发展。

交流项目

AI+无线通信

在无线通信领域,目前已有多个无线通信问题在尝试通过机器学习的方法来提高通信性能,但总体仍受限于以下几个方面:

1)缺乏性能及鲁棒性的保障;

2)需要大量数据来进行模型训练;

3)在应用到具体场景时,线下训练出来的基础模型并不能与实际应用场景匹配。

在此情况下,郑教授提出基于可解释性来分解量化神经网络模型,通过训练出有参数控制以及可插拔的模型,同时整合大量现有已训练好的模型,从而减少匹配应用过程中对数据和算力的要求,并提高训练结果的可重复应用能力。

AI for science

针对生物医药领域现存的蛋白质分子仿真问题,利用深度神经网络模拟量子激发态以量化蛋白质功能,解决量子模拟的准确性和效率问题。

低空通导监网络

随着低空经济的迅猛发展,各类低空飞行器(如无人机、eVTOL等)在城市物流、应急救援、巡检监测等场景中的规模化应用,对底层通信、导航与监测(即“通导监”)基础设施提出了高度差异化且严苛的技术要求。然而,现有单一制式的地面通信网络(如4G/5G、卫星通信或专网系统)在覆盖能力、功能集成度等方面难以全面满足低空空域复杂动态环境下的综合业务需求。

在此背景下,本项目聚焦超高层建筑火灾应急救援场景,联合通信、导航、监测等领域的企业与科研机构,开展面向低空智能空域的“通导监一体化”系统研发。项目通过深度融合5G-A/6G先进通信、北斗高精度定位、多模态感知等核心技术,构建具备自适应组网、智能资源调度与多维态势感知能力的低空通导监融合网络架构。同时,依托真实场景数据,研发可迁移的神经网络模型,最终打造一套可复制、可扩展、支持多行业应用的低空智能通导监基础平台,为低空经济高质量发展提供核心支撑。

学术论文

  • W Yu, KB Letaief, L Zheng, Sensing for Free: Learn to Localize More Sources than Antennas without Pilots. IEEE Journal on Selected Areas in Communications, 2026
  • US Khan, L Liu, S Jere, L Zheng, Y Yi, Configuring RNN Weights for MIMO-OFDM Receive Processing: Informing RNN With Domain Knowledge, IEEE Wireless Communications Letters, 2025
  • R Safavinejad, S Jere, L Zheng, L Liu, Configuring RNN's Recurrent Weights using Domain Knowledge for LTI Approximation, IEEE Signal Processing Letters, 2025
  • S Jere, L Zheng, K Said, L Liu, Towards xAI: Configuring RNN weights using domain knowledge for MIMO receive processing, IEEE Transactions on Wireless Communications, 2025
  • J Xu, L Li, L Zheng, L Liu, Learning to estimate: A real-time online learning framework for MIMO-OFDM channel estimation, IEEE Transactions on Wireless Communications, 2024
  • Xiangxiang Xu and Lizhong Zheng, “Neural Feature Learning in Function Space”, Journal of Machine Learning Research (JMLR), 25(142), 2024
  • J Jon Ryu, Xiangxiang Xu, HS Erol, Yuheng Bu, Lizhong Zheng, Gregory W Wornell, “Operator SVD with Neural Networks via Nested Low-Rank Approximation”, ICML, 2024
  • Shao-Lun Huang, Anuran Makur, Gregory W. Wornell and Lizhong Zheng (2024), “Universal Features for High-Dimensional Learning and Inference”, Foundations and Trends® in Communications and Information Theory: Vol. 21: No. 1-2, pp 1-299.
  • Xiangxiang Xu, Lizhong Zheng: Sequential Dependence Decomposition and Feature Learning, Allerton Conference, 2023
  • Xiangxiang Xu, Lizhong Zheng, Ishank Agrawal: Nerual Feature Learning for Engineering Problems, Allerton Conference, 2023
  • Xiangxiang Xu, Lizhong Zheng: Kernel Subspace and Feature Extraction, IEEE International Symposium on Information Theory (ISIT), 2023
  • Shao-Lun Huang, Xiangxiang Xu, Lizhong Zheng: An Information Theoretic Approach to Unsupervised Feature Selection for High Dimensional Data, IEEE Journal on Selected Areas in Information Theory (JSAIT), Vol. 1, pp157-166, 2020
  • Anuran Makur, Gregory W. Wornell: On Estimation of Modal Decomposition, IEEE International Symposium on Information Theory (ISIT), 2020
  • Shao-Lun Huang, Xiangxiang Xu, Lizhong Zheng, Gregory W. Wornell: A Local Characterization for Wyner Common Information, IEEE International Symposium on Information Theory (ISIT), 2020
  • Shao-Lun Huang, Anuran Makur, Gregory W. Wornell, Lizhong Zheng: On Universal Features for High-Dimensional Learning and Inference, submitted to IEEE Transactions on Information Theory, November, 2019
  • Shao-Lun Huang, Xiangxiang Xu, Lizhong Zheng, Gregory W. Wornell: An Information Theoretic Interpretation to Deep Neural Networks. IEEE International Symposium on Information Theory (ISIT), 2019
  • Hye Won Chung, Brian M. Sadler, Lizhong Zheng, Alfred O. Hero III: Unequal Error Protection Querying Policies for the Noisy 20 Questions Problem. IEEE Trans. Information Theory 64(2): 1105-1131 (2018)
  • Anuran Makur, Lizhong Zheng: Polynomial Singular Value Decompositions of a Family of Source-Channel Models. IEEE Trans. Information Theory 63(12): 7716-7728 (2017)
  • Shao-Lun Huang, Lin Zhang, Lizhong Zheng: An information-theoretic approach to unsupervised feature selection for high-dimensional data. ITW 2017: 434-438
  • Kwang-Cheng Chen, Shao-Lun Huang, Lizhong Zheng, H. Vincent Poor: Communication Theoretic Data Analytics. IEEE Journal on Selected Areas in Communications 33(4): 663-675 (2015)
  • Emmanuel Abbe, Lizhong Zheng: A Coordinate System for Gaussian Networks. IEEE Trans. Information Theory 58(2): 721-733 (2012)
  • Emmanuel Abbe, Lizhong Zheng: Linear Universal Decoding for Compound Channels. IEEE Trans. Information Theory 56(12): 5999-6013 (2010)