About

Zhiqiang He

Zhiqiang He 何志强

Ph.D. researcher in Information Science at the University of Electro-Communications, Tokyo, advised by Prof. Zhi Liu. What I care about is how AI can interact directly with the real world — not an agent that scores well inside a simulator, but one that still works when it meets real users, real networks, and real hardware.

Before that I received my M.S. in Control Science and Engineering from Northeastern University, Shenyang, advised by Prof. Jiao Wang.

That is also why I spent time in industry — an RL algorithms engineer at InspirAI and a research intern at Baidu — where the agents I built had to leave the lab and face real players and real deployments.

Experience

  1. 2022 — 2023

    RL Algorithms Engineer · InspirAI, HangzhouTop Performance Team Video Product

    Built a card-game AI SDK shipped across four production titles; the Landlord agent reached super-human level against top-ranked professional players. A similar (independently developed) approach also powered a Guandan agent, improving its win rate by 6%.

    Dou Dizhu (Landlord) — the flagship title shipping the card-game AI
    Dou Dizhu (Landlord) — the flagship title shipping the card-game AI
    Guandan AI — four agents in a live match (a similar RL approach)
    Guandan AI — four agents in a live match (a similar RL approach)

  2. 2021

    RL Research Intern · Baidu, BeijingSuper Special Offer Video Prototype code

    Single-handedly developed EDA-MAPPO — the full algorithm and its performance gains — and shipped it into a client production environment.

    Deployed system — UAV swarm engaging a naval target (EDA-MAPPO)
    Deployed system — UAV swarm engaging a naval target (EDA-MAPPO)
    light_mappo — open-source prototype (multi-agent PPO)
    light_mappo — open-source prototype (multi-agent PPO)

Awards

  1. 2025–2027 Selected as a JST Next-Generation Researcher (¥2.2M/year stipend plus ¥600K/year research funding).
  2. 2019 Selected as Outstanding Graduate (Top 1%) at East China Jiaotong University.

Service

Peer reviewer for

  • ACM International Conference on Multimedia (ACM MM 2026)
  • IEEE Transactions on Multimedia
  • IEEE Transactions on Network Science and Engineering
  • IEEE Internet of Things Journal
  • IEEE Open Journal of the Computer Society Certificate

Conference volunteer

  • Student Volunteer, IEEE INFOCOM 2026, Tokyo Certificate

Selected Publications

For a complete list of publications, please visit my Google Scholar page.

First / corresponding author

  1. ReForge: Keeping ABR Algorithms Never Finished with Verified Large Language Model Edits

    Zhiqiang He, Zhi Liu

    arXiv preprint, 2026Preprint

    arXiv PDF

  2. NSMA: Neuro-Symbolic Manifold Alignment for Generalizable Adaptive Bitrate Streaming under Texture Shift

    Zhiqiang He, Zhi Liu

    arXiv preprint, 2026Preprint

    arXiv Project PDF

  3. Plasticity-Aware Mixture of Experts for Learning Under QoE Shifts in Adaptive Video Streaming

    Zhiqiang He, Zhi Liu

    IEEE Transactions on Multimedia, 2026 · IF 9.7 · JCR Q1 · CCF-AAccepted

    Paper arXiv PDF Code

  4. Silent Neuron Theory and Plasticity Preservation for Deep Reinforcement Learning in Adaptive Video Streaming

    Zhiqiang He, Zhi Liu

    arXiv preprint, 2025Preprint

    arXiv PDF

  5. A Survey on DRL based UAV Communications and Networking: DRL Fundamentals, Applications and Implementations

    Wei Zhao, Shaoxin Cui, Wen Qiu*, Zhiqiang He*, Zhi Liu, Xiao Zheng, Bomin Mao, Nei Kato

    IEEE Communications Surveys & Tutorials, 2025 · IF 42.8 · JCR Q1

    Paper arXiv PDF

  6. Understanding World Models through Multi-Step Pruning Policy via Reinforcement Learning

    Zhiqiang He, Wen Qiu, Wei Zhao, Xun Shao, Zhi Liu

    Information Sciences, 2024 · IF 8.1 · JCR Q1

    Paper PDF Code

2026

  1. DiPerceiveNet: A bidirectional cross-scale perception network for vehicle re-identification

    Jihao Cai, Zhiqiang He, Zhi Liu, Yangjie Cao

    Pattern Recognition, 2026 · IF 7.6 · JCR Q1

    Paper PDF

  2. Plasticity-Enhanced Multi-Agent Mixture of Experts for Dynamic Objective Adaptation in UAV-Assisted Emergency Communication Networks

    Wen Qiu, Zhiqiang He, Wei Zhao, Hiroshi Masui

    IEEE Internet of Things Journal, 2026 · IF 8.7 · JCR Q1Accepted

    Paper arXiv PDF

2025

  1. Scalable and Reliable Multi-agent Reinforcement Learning for Traffic Assignment

    Leizhen Wang, Peibo Duan, Cheng Lyu, Zewen Wang, Zhiqiang He, Nan Zheng, Zhenliang Ma

    Communications in Transportation Research, 2025 · IF 14.5 · JCR Q1

    Paper PDF Code

2022

  1. Erlang Planning Network: An iterative model-based reinforcement learning with multi-perspective

    Jiao Wang, Lemin Zhang, Zhiqiang He, Can Zhu, Zihui Zhao

    Pattern Recognition, 2022 · IF 8.5 · JCR Q1

    Paper PDF

2018

  1. Control Strategy of Speed Servo Systems Based on Deep Reinforcement Learning

    Pengzhan Chen, Zhiqiang He, Chuanxi Chen, Jiahong Xu

    Algorithms, 2018

    Paper PDF Code