北卡罗来纳州立大学工业与系统工程（ISE）系的研究方向涵盖多个高增长领域，包括高级制造、健康系统工程、人因与人体工程学、供应链与物流，以及系统分析与优化。
The research areas in the Industrial and Systems Engineering (ISE) department at North Carolina State University cover multiple high-growth areas, including advanced manufacturing, health systems engineering, human factors and ergonomics, supply chain and logistics, and systems analysis and optimization. 研究方向包括下一代无线和移动网络系统的优化，机器学习应用，无线和移动网络安全分析与防御，信息物理系统安全与隐私保护，分布式数据共享/分析/发布以保护隐私，鲁棒和隐私保持的机器学习/联邦学习。
Research areas include optimization of next-generation wireless and mobile network systems, machine learning applications, wireless and mobile network security analysis and defense, information-physical system security and privacy protection, distributed data sharing/analysis/publishing for privacy preservation, robust and privacy-preserving machine learning/federated learning. 自然语言处理（NLP）和人工智能（AI），重点关注社会计算，特别是LLM/VLM的社会意识，可解释性和以人为中心的多样性。文章主要发表在顶级NLP和AI会议上，如EMNLP，NAACL，ACL。优先考虑具有计算机科学，信息工程以及相关领域背景的研究经验者。实验室目前得到政府（例如NSF）和工业界（例如Google）项目的支持，经费充足。
Natural Language Processing (NLP) and Artificial Intelligence (AI) with a focus on social computing, particularly social awareness for LLM/VLM, explainability, and human-centric diversity. Publications are mainly in top NLP and AI conferences such as EMNLP, NAACL, ACL. Research background in Computer Science, Information Engineering, and related fields is preferred. The lab is currently supported by government (e.g., NSF) and industry (e.g., Google) projects with sufficient funding. 1. 区块链网络基础设施的安全，比如区块链客户端节点，P2P网络，以及第三方RPC服务的安全。 2. 智能合约安全，比如智能合约分析与漏洞检测，模糊测试等。 3. 区块链应用安全，比如加密货币黑产分析和检测，诈骗合约和流氓地址的检测。
1. Security of blockchain network infrastructure, such as blockchain client nodes, P2P networks, and security of third-party RPC services. 2. Security of smart contracts, such as smart contract analysis and vulnerability detection, fuzz testing, etc. 3. Security of blockchain applications, such as analysis and detection of cryptocurrency black markets, detection of fraudulent contracts and rogue addresses. 人工智能/机器学习 (AI/ML)，生物信息学和计算生物学(Bioinformatics/Computational Biology)（重点关注单细胞分析，多组学分析，空间转录组，癌症研究，智能健康和精准医疗）
Artificial Intelligence/Machine Learning (AI/ML), Bioinformatics/Computational Biology (with a focus on single cell analysis, multi-omics analysis, spatial transcriptomics, cancer research, intelligent healthcare, and precision medicine) 在电动汽车领域的研究兴趣，包括电动汽车驾驶员行为分析、充电基础设施的战略部署、充电时间表的优化以及电动汽车激励政策的设计和评估。也关注自动驾驶汽车技术，着重研究可能促进CAV采用的基础设施，以及为现有CAV技术制定策略，如卡车队列和模块化自动驾驶汽车。
Research interests in the field of electric vehicles, including behavior analysis of electric vehicle drivers, strategic deployment of charging infrastructure, optimization of charging schedules, design and evaluation of electric vehicle incentive policies. Also interested in autonomous vehicle technology, focusing on infrastructure that may facilitate the adoption of CAV, as well as developing strategies for existing CAV technologies, such as truck queues and modular autonomous vehicles. 复杂基础设施网络上的机器学习(主要为强化学习)和优化, 交通网络分析和控制, 网络科学和网络数据分析。导师鼓励跨学科研究。学生的具体研究方向可根据学生的兴趣进行调整。
Machine learning (mainly reinforcement learning) and optimization on complex infrastructure networks, traffic network analysis and control, network science and network data analysis. The advisor encourages interdisciplinary research. Specific research directions can be adjusted according to students' interests. 基于Web的数据中心开发，网络基础设施，数据整理，数据检索，数据共享，分析，软件开发，云计算，机器学习算法，自然语言处理，数据库设计，元数据管理，基于Web的系统开发，科学工作流程，实验数据管理
Web-based Data Hub Development, Cyberinfrastructure, Data Curation, Data Retrieval, Data Sharing, Analytics, Software Development, Cloud Computing, Machine Learning Algorithms, Natural Language Processing, Database Design, Metadata Management, Web-based System Development, Scientific Workflows, Experimental Data Management 应用机器学习（ML）/人工智能（AI）用于工程设计, 工程设计中的人机协作, 数据驱动设计, 多模态学习, 生成建模, 用于生成和评估创新设计的大型语言模型, 人机交互, 人因设计AI, 人-AI混合团队, 可信AI的发展
Applied Machine Learning (ML) / Artificial Intelligence (AI) for engineering design, human-AI collaboration for engineering design, data-driven design, multimodal learning, generative modeling, large language models for generating and assessing innovative designs, human-AI interaction, human factor design of AI, human-AI hybrid teaming, development of trust-worthy AI 智能材料设计（例如刺激响应材料、仿生设计、3D 打印）、个人湿热微环境管理（例如智能纺织品、表面润湿性、热调节、定向液体输送）、智能建筑和环境控制（例如热调节、通风控制、空气质量改善）
Design of smart materials (e.g., stimuli responsive materials, bio-inspired design, 3D printing), Personal thermal and moisture management (e.g., smart textiles, surface wettability, thermal regulation, directional liquid transport), Smart building and environment control (e.g., thermal regulation, ventilation control, air quality improvement) 多相机姿势估计与校准，鲁棒特征匹配和运动结构（SfM）方法，稀疏相机校准，相对姿势估计，具有遮挡和/或很少重叠的相机校准，逼真的3D重建和新颖视图合成，体积和表面3D重建和视图合成，稀疏视图重建和视图合成，场景重建和视图合成，从不同外观，降级图像，短暂遮挡等方面，大规模重建和视图合成，来自卫星图像的数字表面重建，逼真而准确的新颖视图合成和遍历
Multi Camera Pose Estimation and Calibration, Robust feature matching and Structure-from-Motion (SfM) methods, Sparse camera calibration, Relative pose estimation, Camera calibration with occlusion and/or little overlap, Photo-realistic 3D Reconstruction and Novel View Synthesis, Volumetric and surface 3D reconstruction and view synthesis, Sparse view reconstruction and view synthesis, Scene reconstruction and view synthesis from varying appearances, degraded images, transient occlusions, Large scale reconstruction and view synthesis, Digital surface reconstruction from satellite imagery, Photo realistic and accurate novel view synthesis and walk-through 神经生物学，RNA生物学，免疫学，使用超分辨率荧光成像，单分子探测和高通量基因组成像等前沿生物成像分析技术研究与神经退行性疾病和癌症等相关的疾病。实验室的研究项目高度跨学科，涉及化学、物理、生物学和工程学。
Neurobiology, RNA biology, Immunology, diseases related to neurodegeneration and cancer using cutting-edge biological imaging analysis techniques including super-resolution fluorescence imaging, single-molecule detection, and high-throughput genome imaging. The research projects in the lab are highly interdisciplinary, involving chemistry, physics, biology, and engineering. 研究气候变化背景下自然灾害中对沿岸过程的影响，如极端天气事件引发的波浪和风暴潮，以及滑坡和地震活动触发的海啸。利用现有的数值模型评估这些自然灾害对沿岸居民的相关风险。开发基于机器学习的工具以提高数值模型的精确性及计算效率。利用机器学习的工具分析灾害风险。
Research on the impacts of coastal processes in natural disasters under the background of climate change, such as waves and storm surges triggered by extreme weather events, tsunamis triggered by landslides and seismic activity. Using existing numerical models to assess the relevant risks of these natural disasters to coastal residents. Developing machine learning-based tools to improve the accuracy and computational efficiency of numerical models. Analyzing disaster risks using machine learning tools. 研究方向包括平面介电常数和受限水的电导率，原子尺度约束下的毛细凝结，受限水的异常低介电常数，无标记识别单个介电性纳米粒子和病毒，通过原子尺度精度制备的毛细管进行的分子传输。
Research areas include in-plane dielectric constant and conductivity of confined water, capillary condensation under atomic-scale confinement, anomalously low dielectric constant of confined water, label-free identification of single dielectric nanoparticles and viruses, molecular transport through capillaries made with atomic-scale precision. Fully funded research positions in collaboration with industrial partners, including Philips and Canon, focusing on knowledge graphs, ontologies, and the semantic web, integrated with natural language processing and large language models. For more details, visit Zorro Project. Co-supervision is provided by Dr. Stefan Schlobach, head of the department at VU Amsterdam.
My research goal is to advance the capabilities of AI agents to interact with the physical world, with a special emphasis on leveraging generative machine learning approaches for world modeling, iterative reasoning, and effective decision-making. I am particularly interested in developing multi-modal generative models that span natural language, images, videos, and 3D, ensuring these models are Efficient, Flexible, Scalable, and Knowledgeable. By bridging foundational machine learning research with real-world applications, my research seeks to create robust AI systems capable of operating seamlessly in complex and dynamic environments.
图机器学习，图基础模型，图神经网络，数据为中心的人工智能，病态数据问题发现与缓解如不平衡问题，拓扑问题，数据稀少问题，可信人工智能如可解释性，公平性，不确定性等，科学/基础设施/信息检索/网络安全的图人工智能与机器学习应用
Machine Learning on Graphs, Graph Foundational Models, Graph Neural Networks, Data-centric AI, Data-quality Issues Discovery and Mitigation such as Imbalance Issue, Topology Issue, Limited Data Issue, Trustworthy AI such as Explanability, Fairness, Uncertainty, etc., Graph AI for Science/Infrastructure/Information Retrieval/Cybersecurity 大语言模型（LLM）训练，微调，推理; DNN/CV/NLP的高效训练和推理; ML System (efficient large scale training framework); AI算法-硬件协同设计（GPU）
Large Language Model (LLM) training, fine-tuning, inference; Efficient training and inference for DNN/CV/NLP; ML System (efficient large-scale training framework); AI algorithm-hardware Co-design (GPU) Computational Condensed Matter Physics. My research focuses on developing and applying density functional theory (DFT) simulations coupled with theoretical models to design and predict diverse nanomaterials and quantum materials and reveal their novel electronic, topological, magnetic, optical, and phononic properties. My recent research interests include, but not limited to, the study of topological materials, chiral materials, superconductors, and integrating machine learning in materials discovery. I am actively looking for PhD students. If you are interested, please email me your CV.
1. Computational Biology, especially in a trustworthy manner like learning biological relevant signals from different species, different populations, different experiment protocols or devices, especially applications genomics and Alzheimer’s disease. 2. Large Language Model as Agents and Trustworthy AI: developing LLM-based agents to fully automate models' completion of complicated tasks, including testing AI's security. 3. The combination of the above two.
可信机器学习（如稳健性，公平性，安全性，因果关系，可解释性，可迁移性）；生成式人工智能（如大型语言模型，大型多模态模型，扩散模型）；因果推断的机器学习（如复杂治疗方法，持续因果推断，网络上的因果推断）；视觉智能（如域自适应，领域泛化，场景图生成，密集预测）；科学、健康和教育领域的人工智能
Trustworthy Machine Learning (e.g., robustness, fairness, safety, causality, explainability, transferability); Generative Artificial Intelligence (e.g., large language models, large multimodal models, diffusion models); Machine Learning for Causal Inference (e.g., complex treatments, continual causal inference, causal inference on networks); Visual Intelligence (e.g., domain adaptation, domain generalization, scene graph generation, dense prediction); AI for Science, Health, and Education 人工智能/机器学习，网络，安全，体系结构，无线网络，网络安全，深度强化学习，可信人工智能，可解释人工智能，安全智能计算，模型压缩，人工智能效率，人工智能安全，人工智能系统与软硬件协同设计
AI/ML, Networking, Security, Architecture, Wireless Network, Cyber Security, Deep Reinforcement Learning, Trustworthy AI, Explainable AI, Secure and Intelligent Computing, Model Compression, AI Efficiency, AI Security, AI System and SW/HW Co-design 1. Memory system performance optimization 2. In-storage computing and near-storage computing 3. Accelerating emerging applications (such as artificial intelligence, graph algorithms, etc.) 机器学习（深度学习，强化学习，迁移学习），统计学（因果推断，贝叶斯统计，空间统计，高维统计，时间序列分析，存活分析，元分析，图模型），生物统计和生物信息，精算科学，应用数学，临床试验设计等
Machine Learning (Deep Learning, Reinforcement Learning, Transfer Learning), Statistics (Causal Inference, Bayesian Statistics, Spatial Statistics, High-dimensional Statistics, Time Series Analysis, Survival Analysis, Meta-analysis, Graphical Models), Biostatistics and Bioinformatics, Actuarial Science, Applied Mathematics, Clinical Trial Design, etc. 1. 移植物抗宿主病（GVHD）的生物学：调节和炎症信号、GvHD发病率和死亡率的生物标志物。2. 骨髓源性抑制细胞（MDSCs）的生物学：调节分化、迁移、免疫抑制功能、代谢以及治疗策略。3. 多发性骨髓瘤肿瘤微环境的免疫学：免疫系统相互作用、控制恶性浆细胞增殖、开发新的治疗策略。
1. Biology of Graft-versus-host disease (GVHD): regulatory and inflammatory signaling, biomarkers for GvHD morbidity and mortality. 2. The biology of myeloid-derived suppressor cells (MDSCs): regulation of differentiation, trafficking, immunosuppressive function, metabolism, and therapeutic strategies. 3. Immunology of tumor microenvironment in multiple myeloma: immune system interactions, control of malignant plasma cell proliferation, development of new treatment strategies. 我的研究主要集中在软件工程、网络安全和数据科学的交叉领域，涉及社会技术方面，以及对不同类型软件构件（例如代码、执行跟踪、错误报告、问答帖子和开发者网络）的分析及它们之间的相互作用。我们特别感兴趣将被动软件工程数据转化为自动化工具，以提高系统的可靠性、安全性和性能，提高开发者的生产力，并为决策者产生新的见解。
My research primarily works at the intersection of software engineering, cybersecurity and data science, encompassing socio-technical aspects, and analysis of different kinds of software artifacts (e.g., code, execution traces, bug reports, Q&A posts, and developer networks) and the interplay among them. We are particularly interested in transforming passive software engineering data into automated tools that can improve system reliability, security, and performance, increase developer productivity, and generate new insights for decision-makers. 梅胜林实验室专注于肿瘤微环境在肿瘤进展和转移过程中的重塑和调控机制。目前的研究方向包括开发整合多模态数据的计算方法，研究肿瘤微环境的环境依赖性重塑，识别肿瘤微环境中的新调控因子，并探索肿瘤器官特异性转移与多组学数据整合。
The Mei Laboratory focuses on the remodeling and regulatory mechanisms of the tumor microenvironment during tumor progression and metastasis. Current research directions include developing computational methods for integrating multi-modal data, investigating context-dependent remodeling of the tumor microenvironment, identifying novel regulators within the tumor microenvironment, and exploring tumor organ-specific metastasis with multi-omics data integration. 该研究项目致力于开发和应用机器学习/深度学习方法进行生物医学图像分析和预测。涉及利用先进的神经影像技术进行人工智能驱动的神经生物标志物发现，用于脑机交互和精准医学。
The research project focuses on developing and applying machine learning/deep learning methods for biomedical imaging analysis and prediction. It involves artificial intelligence (AI)-driven neural biomarker discovery using advanced neuroimaging techniques for brain-computer interaction and precision medicine. 设计、开发和应用先进功能膜材料，实现水—能源—食物系统的循环经济。研究包括材料创新、过程设计、理论模型和人工智能，用于开发新型膜材料和膜过程，用于饮用水净化、新型污染物处理、废弃物回收和资源再利用。
Design, development, and application of advanced functional membrane materials for achieving circular economy in the water-energy-food system. Research includes material innovation, process design, theoretical models, and artificial intelligence for developing new membrane materials and membrane processes for drinking water purification, novel pollutant treatment, waste recovery and resource reuse. 细胞应激反应的分子机制，重点关注线粒体应激、内质网（ER）应激和下游的整合应激反应（ISR）。采用跨学科方法，包括结构生物学（冷冻电镜/Cryo-ET）、生物化学（蛋白质表征）和细胞生物学（荧光成像）来开展研究。旨在调节生理和病理条件下的应激反应途径，为人类疾病如神经退行性疾病、代谢紊乱和癌症等开发新疗法。与疾病相关小鼠模型领域的专家合作，以补充机制性研究成果。
Cell stress response molecular mechanisms, focusing on mitochondrial stress, endoplasmic reticulum (ER) stress, and downstream integrated stress response (ISR). Interdisciplinary approaches including structural biology (cryo-EM/Cryo-ET), biochemistry (protein characterization), and cell biology (fluorescence imaging) are utilized. Aim to regulate stress response pathways under physiological and pathological conditions for the development of novel therapies for human diseases like neurodegenerative diseases, metabolic disorders, and cancer. Collaborations with experts in disease-relevant mouse models field to complement mechanistic research findings. 基于新型材料系统的电子和光子器件的设计、制造和表征，包括低维材料、相变材料和超晶格结构。通过光谱和显微技术理解新兴材料平台中的能量转移机制。
Design, manufacturing, and characterization of electronic and photonic devices based on novel material systems, including low-dimensional materials, phase-change materials, and superlattice structures. Understanding energy transfer mechanisms in emerging material platforms through spectroscopy and microscopy techniques. 软件可靠性，软件安全，系统安全，软件工程，Deepfake检测，生成式AI，机器学习，计算机视觉，自动驾驶，智能交通，分布式量子计算和网络，边缘计算，联邦学习，AI系统与应用，形式化验证，类型论，编程语言，移动计算，数字医疗，无线感知
Software Reliability, Software Security, System Security, Software Engineering, Deepfake Detection, Generative AI, Machine Learning, Computer Vision, Autonomous Driving, Intelligent Transportation, Distributed Quantum Computing and Networking, Edge Computing, Federated Learning, AI Systems and Applications, Formal Verification, Type Theory, Programming Languages, Mobile Computing, Digital Healthcare, Wireless Sensing 边缘人工智能在移动设备、AR、可穿戴设备、机器人和物联网中的应用；高效多模态大语言模型和生成式人工智能在特定领域的应用；人工智能在无线网络、视频/音频分析和智能感知系统中的应用；大语言模型和生成式人工智能的系统架构；联邦学习及其系统实现；面向健康科技的人机交互和移动健康（mHealth），如数字心理健康、智能助听器、手语翻译、药品识别等
Edge AI applications in mobile devices, AR, wearable devices, robots, and IoT; Efficient multimodal large language models and generative AI with applications in specific domains; AI applications in wireless networks, video/audio analysis, and intelligent perception systems; System architecture of large language models and generative AI; Federated Learning (FL) and its system implementation; Human-computer interaction and mobile health (mHealth) for health technology, such as digital mental health, smart hearing aids, sign language translation, drug identification, etc. 机器学习理论（例如，强化学习理论，赌博学习理论和生成式人工智能），凸优化和非凸优化算法设计与分析，系统资源分配理论，无线通信网络性能分析，边缘计算和云计算，智能电网控制理论
Machine Learning Theory (e.g., Reinforcement Learning Theory, Bandit Learning Theory, and Generative AI), Convex and Non-convex Optimization Algorithm Design and Analysis, System Resource Allocation Theory, Wireless Communication Network Performance Analysis, Edge Computing and Cloud Computing, Smart Grid Control Theory (1) AI for Science in chemistry and material science, including machine learning interatomic potentials, generative models, and reaction networks development & applications (2) Atomistic simulations with statistical mechanics & first-principles calculations in disordered/amorphous/interfacial systems (3) Computational modeling of complex materials for renewable energy applications (e.g., Li/Na-ion battery cathodes/electrolytes/interfaces)
GNN, LLM, IR, RecSys, Time Series: Graph Neural Networks. Text-attributed graph and multimodal graph with (M)LLMs; Graph foundation models; Robustness and scalability of graph learning (OOD, test-time, condensation, distillation, graph-MLP etc.). Recommendations and Information Retrieval. RL for recommendations; Rich side-information in recommendations with (M)LLMs; Domain-specific IR and document understanding with LLMs. Time Series. Foundation models for time series; LLMs for time series; Time series for cross-disciplinary applications.
机器学习，重点关注将交互式机器学习与基础模型相连接。具体研究方向包括AI安全与对齐，基于LLM的交互式学习代理和系统，大型模型的高效训练和推断，大型模型的测试和评估系统，以及变压器/注意力/交互式ML的理论基础。
Machine learning, with a focus on connecting interactive ML with foundation models. Specific research directions include AI safety and alignment, LLM-based interactive learning agents and systems, efficient training and inference for large models, testing and evaluation systems for large models, and theoretical foundations of transformer/attention/interactive ML. 机器学习与人工智能（可信的机器学习，学习增强算法，高效的大型人工智能模型等），决策智能与优化（强化学习，在线优化/控制，在线资源分配，在线匹配等），信息物理系统与关键设施（数据中心、能源系统和智能城市应用中的资源管理等），云/边缘计算与网络（能量收集系统，边缘人工智能，分布式计算等）
Machine Learning and Artificial Intelligence (trustworthy ML, learning-augmented algorithms, efficient large AI models, etc.), Decision Intelligence and Optimization (reinforcement learning, online optimization/control, online resource allocation, online matching, etc.), Cyber Physical Systems and Critical Infrastructures (resource management in data centers, energy systems, and smart city applications, etc.), Cloud/Edge Computing and Networks (energy harvesting systems, edge AI, distributed computing, etc.) 硬件安全，包括但不限于汽车系统安全，硬件安全，侧信道攻击和算法，木马检测，故障攻击和检测，后量子密码算法，PUF安全与应用，以及机器学习在硬件安全领域的应用和扩展。
Hardware Security, including but not limited to Automotive System Security, Hardware Security, Side-Channel Attacks and Algorithms, Trojan Detection, Fault Attacks and Detection, Post-Quantum Cryptography Algorithms, PUF Security and Applications, and the Application and Extension of Machine Learning in Hardware Security. 1) 计算机体系结构设计，如Compute Express Link and GPU for performance and sustainablity。2) 硬件安全，如hardware-based attacks and defenses on GPU and System-on-Chips。3) 机器学习系统，ML systems on serverless computing,  personalization and retrieval-augmented generation based on Large language models。
1) Computer Architecture Design, such as Compute Express Link and GPU for performance and sustainability. 2) Hardware Security, such as hardware-based attacks and defenses on GPU and System-on-Chips. 3) Machine Learning Systems, focusing on ML systems on serverless computing, personalization, and retrieval-augmented generation based on Large language models. 开发和应用新颖可信的机器学习（ML）和人工智能（AI）解决方案，以解决医疗挑战和新兴生物医学问题。利用包括电子健康记录、临床记录、信号或多组学数据在内的大规模电子健康数据，跨越各种医疗领域。专注于医院急诊、住院和重症医学研究，涉及无监督学习聚类、监督学习、时间数据建模、临床记录分析和模型公平解释性。
Development and application of novel, trustworthy machine learning (ML) and artificial intelligence (AI) solutions for healthcare challenges and emerging biomedical problems. Working with large-scale electronic health data including electronic health records, clinical notes, signals, or multi-omics data across various healthcare domains. Focus on hospital emergency, inpatient, and critical care medicine research involving unsupervised learning clustering, supervised learning, temporal data modeling, clinical note analysis, and model fairness interpretability. 新架构的编译器和编程系统, 量子机器学习和量子化学的机器学习, 量子电路优化, 量子纠错, 表面码, 晶格手术, LDPC码, 强化学习和扩散模型用于生成量子电路构建, 量子位映射, 脉冲优化, 量子最优控制和误差缓解, 波色计算, 混合连续变量(CV)离散变量(DV)量子系统, 量子近似优化算法(QAOA), 变分量子本征求解器(VQE)和量子傅里叶变换(QFT), 存储器局部性优化, 计算重排序和数据布局变换, 以及共享存储器层次结构, GPU, 信息流跟踪, 寄存器分配, 二进制解码/仪器化, 图分区模型及其在GPU局部性优化中的应用, 马尔科夫链, 排队模型和跟踪拟合
Compiler and programming systems for emerging architectures, Quantum machine learning and machine learning for quantum chemistry, Quantum circuit optimization, quantum error correction, surface code, lattice surgery, LDPC codes, Reinforcement learning and diffusion models for generative quantum circuit construction, Qubit mapping, pulse optimization, quantum optimal control, and error mitigation, Bosonic computing, hybrid continuous-variable (CV) discrete-variable (DV) quantum systems, Quantum approximate optimization algorithms (QAOA), variational quantum eigensolvers (VQE), and quantum Fourier transformation (QFT), Memory locality optimization, computation reordering and data layout transformation, and shared memory hierarchies, GPUs, information flow tracking, register allocation, and binary decoding/instrumentation, Graph partition models and their application for GPU locality optimization, Markov chains, queueing models, and trace fitting 以人为中心的人工智能，以社会为本的人工智能，可信人工智能，XAI，生成式人工智能，多模态大型语言模型，虚假信息检测与解释，推荐系统，众包，众感，人工智能集体智慧，联邦学习，边缘计算，物联网，隐私，鲁棒性，FATE（公平性，责任性，透明性，伦理学）
Human-centered AI, AI for Social Good, Trustworthy AI, XAI, Generative AI, Multimodal Large Language Model, Misinformation Detection and Explanation, Recommender Systems, Crowdsourcing, Crowdsensing, Human-AI Collective Intelligence, Federated Learning, Edge Computing, Internet of Things (IoT), Privacy, Robustness, FATE (Fairness, Accountability, Transparency, Ethics) 1. 创伤后应激障碍的机制研究与疗法开发; 2. 环境应激因素与创伤后应激障碍的关系; 3. 脑损伤（缺血性中风、创伤性脑损伤、新生儿缺血缺氧）的病理生理学与干预策略; 4. 神经退行性疾病（阿尔茨海默症）的机制研究与治疗方法开发。
1. Mechanisms and therapy development of post-traumatic stress disorder; 2. Relationship between environmental stressors and post-traumatic stress disorder; 3. Pathophysiology and intervention strategies of brain injuries (ischemic stroke, traumatic brain injury, neonatal hypoxic-ischemic encephalopathy); 4. Mechanisms and treatment methods development of neurodegenerative diseases (Alzheimer's disease). 1) 人工智能和机器学习, 2) 大数据处理和算法, 3) 跨学科合作方向：健康大数据和智慧医疗, 4) 跨学科合作方向：智慧医疗设备和机器学习算法, 5) 高效计算和软硬件设计
1) Artificial Intelligence and Machine Learning, 2) Big Data Processing and Algorithms, 3) Interdisciplinary Cooperation: Health Big Data and Smart Healthcare, 4) Interdisciplinary Cooperation: Smart Medical Devices and Machine Learning Algorithms, 5) Efficient Computing and Hardware Design 计算机系统，体系结构，包括基于GPU和FPGA的机器学习加速器设计，存储系统，3D芯片设计。研究涉及到尖端的微处理器设计与开发，包括神经网络加速器，片上网络，GPU，3D存储，非易失性存储，LLM。应用和前景广泛。也欢迎具有电子工程背景的学生申请。该研究组拥有良好的科研环境和合作团队。加入该研究组的博士生在第一年已经发表了两篇在知名科研会议上的论文（DAC为第一作者，ISVLSI）。
Computer Systems, System Architecture, including GPU and FPGA-based machine learning accelerator design, storage systems, 3D chip design. Research involves cutting-edge microprocessor design and development, including neural network accelerators, on-chip networks, GPU, 3D storage, Non-volatile storage, LLM. Wide range of applications and prospects. Students with backgrounds in electronic engineering are also welcome to apply. The research group has a good research environment and collaborative team. PhD students who joined the group have already published two papers in well-known research conferences in their first year (DAC as first author, ISVLSI). 开发可访问的诊断工具，整合化学/生物学/医学、微-纳米技术和消费电子技术，以解决癌症、传染病、心脑血管疾病和神经系统疾病等疾病。利用微纳技术、机器学习、合成生物学、微流控和仿生学。
Accessible diagnostic tools, integrating chemistry/biology/medicine, micro- and nanotechnology, and consumer electronics to address diseases like cancer, infectious diseases, cardiovascular and neurological disorders. Utilizing micro/nano-technologies, machine learning, synthetic biology, microfluidics, and biomimetics. 利用质谱、基因编辑和下一代测序技术研究环境暴露和内源代谢诱导的DNA损伤如何导致癌症和与衰老相关的疾病。在蛋白质组学、化学生物学、合成化学、基因编辑和生物信息学等领域接受跨学科培训。
Leveraging mass spectrometry, gene editing, and next-generation sequencing techniques to study how DNA damage induced by environmental exposure and endogenous metabolism contribute to cancer and aging related diseases. Interdisciplinary training in proteomics, chemical biology, synthetic chemistry, gene editing, and bioinformatics. 材料科学与工程，机械工程，电气工程，化学工程。具体研究方向为基于电池的长时能量存储在系统和材料层面，包括电池管理系统，电池数据分析，电池材料/电池单体表征工具以及电池组设计。
Material Science and Engineering, Mechanical Engineering, Electrical Engineering, Chemical Engineering. Specifically focusing on battery-based long duration energy storage at system and material level, including battery management systems, battery data analysis, battery material/cell characterization tools, and design of battery packs. 安全人机交互/协作，多智能体学习和规划，机器人学习，具身人工智能，机器人和自主系统的基础模型，安全和交互自主驾驶和机器人导航，值得信任的人工智能/机器学习，深度强化学习和控制，智能交通系统与智慧城市
Safe Human-Robot Interaction / Collaboration, Multi-Agent Learning & Planning, Robot Learning, Embodied AI, Foundation Models for Robotics and Autonomous Systems, Safe and Interactive Autonomous Driving & Robot Navigation, Trustworthy AI/ML, Deep Reinforcement Learning & Controls, Intelligent Transportation Systems & Smart City 能源传输的计算建模和实验表征，热声子与其他能量载体（电子、光子、磁子）的相互作用，用于可再生能源转换、微电子制冷、空间和建筑技术、增材制造、生物医学工程、量子计算的材料设计和发现
Computational modeling and experimental characterization of energy transport, thermal phonons interactions with other energy carriers (electrons, photons, magnons), material design and discovery for applications in renewable energy conversion, microelectronics cooling, space and building technologies, additive manufacturing, biomedical engineering, quantum computing 生物医学工程、计算机科学、计算机工程、环境工程、电子工程、森林学、数学/统计、机械工程、航天工程、核工程、纺织科学/纺织工程、英语、国际研究、商科/管理、心理学、传播学、护士等
Biomedical Engineering, Computer Science, Computer Engineering, Environmental Engineering, Electrical Engineering, Forestry, Mathematics/Statistics, Mechanical Engineering, Aerospace Engineering, Nuclear Engineering, Textile Science/Textile Engineering, English, International Studies, Business/Management, Psychology, Communication, Nursing, among others 增强型语言模型中的工具/流程/任务的自动化，增强型语言模型中的决策/推理的可靠性，可扩展/高效/安全的增强型语言模型系统，生成式AI中的数据优化/增强/选择，生成式AI中的算法-硬件协同设计
Tool/Chain/Task Automation in ALM, Decision/Reasoning Reliability in ALM, Scalable/Efficient/Safe ALM Systems, Data Optim/Augment/Select for Generative AI, Algo-Hardware Co-design for Generative AI 1. 基于类脑器件（包含但不限于忆阻器&硅光）的机器学习/深度学习算法算法以及电/光路硬件设计。2. 结合类脑硬件（包含但不限于Event-based Camera）的机器学习/深度学习算法。3. 生物神经网络启发的新机器学习/深度学习算法（包含但不限于Spiking Neural Network & Hopfield Neural Network & Reservoir Computing）。4. 也欢迎同学沟通，尊重每一个新的idea或方向，希望可以头脑风暴一些有趣的课题。
1. Machine learning/deep learning algorithms and electrical/optical circuit hardware design based on brain-like devices (including but not limited to memristors & silicon photonics). 2. Machine learning/deep learning algorithms combined with brain-like hardware (including but not limited to Event-based Camera). 3. New machine learning/deep learning algorithms inspired by biological neural networks (including but not limited to Spiking Neural Network, Hopfield Neural Network, Reservoir Computing). 4. Open to new ideas and directions, brainstorming for interesting topics. 研究兴趣主要涵盖尾矿管理的各个方向，结合理论研究和实验研究。研究领域包括但不限于：尾矿沉降和固结，尾矿过滤性能，尾矿的流变学、岩土力学和力学行为，膏体充填，尾矿库稳定性，尾矿管理中的CO2固定。
Research interests mainly cover various aspects of tailings management, combining theoretical and experimental research. Research areas include but are not limited to: Sedimentation and consolidation, Filtration performance of thickened tailings, Rheological, geotechnical, and mechanical behaviors of tailings, Paste tailings backfill, Stability of tailings storage facilities, CO2 sequestration in tailings management. 应用密码学，包括但不限于密码分析、加密数据库、分布式系统中的密码学应用。欢迎对应用密码学其他分支感兴趣的学生。
Applied Cryptography, including but not limited to cryptanalysis, encryption databases (based on searchable encryption, private information retrieval, fully-homomorphic encryption constructions), and cryptographic applications in distributed systems. Other branches of applied cryptography are also welcomed. 3. 量子计算在信息安全和机器学习上的应用
1. Quantum computing complexity and foundational issues in quantum cryptography 2. Validation of quantum experimental devices 3. Applications of quantum computing in information security and machine learning 无线网络，包括通信、网络、安全和感知，专注于利用机器学习技术提升无线网络在能效、实时响应、可扩展性、灵活性和安全性等关键性能方面的表现。未来研究方向包括下一代无线网络（例如，在 6G 网络和自动驾驶场景下基于多模态传感器融合的毫米波信道预测、目标跟踪、波束成形和网络安全）以及无源物联网（包括跨协议、跨介质、超低功耗通信、联网及其安全等）。
Wireless networks, including communication, networking, security, and sensing, with a focus on utilizing machine learning techniques to enhance the performance of wireless networks in key aspects such as energy efficiency, real-time responsiveness, scalability, flexibility, and security. Future research directions include next-generation wireless networks (e.g., millimeter-wave channel prediction based on multimodal sensor fusion in 6G networks and autonomous driving scenarios, target tracking, beamforming, and network security) and passive Internet of Things (including cross-protocol, cross-medium, ultra-low-power communication, networking, and security). 开发小分子荧光探针及单分子荧光成像技术。具体寻找有机合成荧光探针、荧光光谱学、共聚焦显微镜成像、细胞染色、单分子超分辨成像、光学平台搭建、图像处理和机器学习背景的候选人。
Development of small molecule fluorescent probes and single molecule fluorescence imaging techniques. Specifically looking for candidates with backgrounds in organic synthesis of fluorescent probes, fluorescence spectroscopy, confocal microscopy imaging, cell staining, single-molecule super-resolution imaging, optical platform construction, image processing, and machine learning. 研究领域包括材料物理和科学，重点关注能量材料和界面，生物聚合物的结构-性质，以及功能团簇。所使用的方法包括多尺度建模，第一性原理计算，分子动力学模拟，力场分子动力学模拟，机器学习等。该课题组与其他实验研究小组合作。
Research areas include materials physics and science focusing on energy materials and interfaces, structure-property of biopolymers, and functional clusters. Methods utilized range from multiscale modeling, first-principles calculations, molecular dynamics simulations, force field molecular dynamics simulations, to machine learning. The group collaborates with other experimental research groups. 南佛罗里达大学生物热研究实验室（BTR）致力于研究生物系统在热处理期间/之后的传热和生物学变化，重点关注各种生物系统的低温保存。具体研究领域包括低温保存系统设计，海洋生物低温保存，冷冻保护剂设计以及冷冻保存的供应链。鼓励对传热基础，各种生物体的冷冻保存，微纳米制造和微流体学感兴趣的申请人申请。
Research in the Bio-Thermal Research Lab (BTR) at the University of South Florida focuses on heat transfer and biological changes in biological systems during/after thermal manipulations, with a focus on cryopreservation of various biological systems. Specific research areas include cryopreservation system design, marine life cryopreservation, cryoprotectant agent design, and supply chain for cryopreservation. Candidates with interests in fundamental heat transfer, cryopreservation of various organisms, micro/nanofabrication, and microfluidics are encouraged to apply. 保护用户隐私，从不断发展的传感器和AI/ML时代的侧信道问题中，对声学、激光、电磁干扰等物理信号对计算机视觉、语音处理等感知系统的安全影响进行建模，为安全和医疗保健设计新颖的传感软件和硬件系统，Android、AR/VR等新兴移动计算平台以及AI DeepFake多媒体中与感知相关的安全和隐私问题。
Protecting user privacy from evolving sensors and side-channel problems in the age of AI/ML, Modeling the security impact of physical signals such as acoustics, lasers, and electromagnetic interference on computer vision, speech processing, and other sensing systems, Designing novel sensing software and hardware systems for security and healthcare, Sensing-related security and privacy problems in Android, AR/VR, and other emerging mobile computing platforms, as well as AI DeepFake multimedia. 机器学习, 医疗数据异质性, 多模态对比学习, 时间序列表示学习, 嵌入式系统设计, 生物信号感知与分析, 可穿戴设备与数字健康, 临床结果, 实时健康监测系统, 心血管数字孪生
Machine learning, Medical data heterogeneity, Multimodal contrastive learning, Time-series representation learning, Embedded System Design, Biosignal sensing and analysis, Wearable and digital health, Clinical Outcome, Real-time Health Monitoring Systems, Cardiovascular Digital Twin 研究方向主要集中在移动和物联网安全、系统安全、软件安全等领域。具体研究包括移动无线网络漏洞、系统安全、软件安全、虚拟化、可信执行环境（TEE）、逆向工程、漏洞检测和二进制加固等。
Research areas mainly focus on Mobile and IoT Security, System Security, and Software Security. Specific research includes vulnerabilities in mobile wireless networks (e.g., 5G/6G), system security, software security, virtualization, Trusted Execution Environment (TEE), reverse engineering, vulnerability detection, and binary hardening. 程光亮：1. 大语言-视觉模型相关研究及应用。2. 机器人感知和智能抓取。3. 图像/视频生成、鉴伪以及安全性等研究方向。4. 智能遥感和医疗方向。彭蓓：任何与强化学习相关的方向，包括单智能体强化学习，多智能体强化学习，和Human-in-the-loop强化学习。联合方向：计算机视觉+强化学习。
Dr. Guangliang Cheng: 1. Large-scale language-vision models and applications. 2. Robot perception and intelligent grasping. 3. Research on image/video generation, authentication, and security. 4. Intelligent remote sensing and medical directions. Dr. Bei Peng: Any direction related to reinforcement learning, including single-agent reinforcement learning, multi-agent reinforcement learning, and human-in-the-loop reinforcement learning. Joint direction: Computer vision + reinforcement learning. 热辐射，纳米光子学，能量转换，电子学，等离子体学，二维材料，数值设计，辐射制冷，太阳能加热，热光子学，等离子体光子学，光子介导热引擎，制冷器，电子和光子电路，辐射特性，增强辐射传热，光子化学势，光发射
Thermal radiation, nanophotonics, energy conversion, electronics, plasmonics, 2D materials, numerical design, Radiative cooling, solar heating, Thermal photonics, plasmonics, Photon-mediated heat engines, refrigerators, Electronic and photonic circuits, Radiative properties, Enhanced radiative heat transfer, Photon chemical potential, light emission 智能健康的机器学习，遥感的机器学习，联邦学习，自监督学习，边缘云计算，集成感知和通信，计算机视觉，语义通信，图学习，压缩感知，人工智能与遥感，人工智能与无线通信，人工智能与智能健康
Machine Learning for Smart Health, Machine Learning for Remote Sensing, Federated Learning, Self-supervised Learning, Edge-cloud computing, Integrated Sensing and Communication, Computer Vision, Semantic Communication, Graph Learning, Compressive Sensing, AI & Remote Sensing, AI & Wireless Communications, AI & Smart Health 研究方向包括强化学习（RL）和多臂老虎机，联邦化/去中心化学习（用于基础模型），具有人类反馈的RL以及与大型语言模型（LLMs）对齐，以及下一代网络系统中的资源管理/优化（例如，无线网络，边缘/云计算系统等）
Research areas include reinforcement learning (RL) and multi-armed bandits, federated/decentralized learning (for foundation models), RL with human feedback and alignment with Large Language Models (LLMs), and resource management/optimization in Next-generation networked systems (e.g., wireless networks, edge/cloud computing systems, etc) 多模态人工智能、自动驾驶、可信和高效机器学习等
深度学习/机器学习算法，整合图像数据和组学数据（尤其是空间转录组和单细胞转录组），器官发育和病理学，基于foundation model（如ChatGPT，LLaMA）的算法开发与建模，对单/多模态生物数据的理解，细胞分化的动态定量模型构建。
Deep learning/machine learning algorithms, integration of imaging data and omics data (especially spatial transcriptomics and single-cell transcriptomics), organ development and pathology, algorithm development and modeling based on foundation models (such as ChatGPT, LLaMA), understanding of single/multi-modal biological data, construction of dynamic quantitative models of cell differentiation. 5. 各类与机器学习以及自动控制相关的研究
1. Security verification of machine learning methods (such as neural networks) 2. Evaluation and improvement of neural network compression methods 3. Data-driven modeling and control 4. Application of LLM in control systems 5. Research related to machine learning and automatic control 不确定性量化、科学机器学习、风险与可靠性评估、灵敏度分析、贝叶斯推断、数据同化和模型验证应用于防灾减灾（地震、台风、火灾）、复合材料、航空航天、计算力学
Uncertainty Quantification, Scientific Machine Learning, Risk and Reliability Assessment, Sensitivity Analysis, Bayesian Inference, Data Assimilation, Model Verification applied in Disaster Prevention and Mitigation (earthquakes, typhoons, fires), Composite Materials, Aerospace, Computational Mechanics 多模态基础模型, 大型语言模型, 视觉-语言基础模型, 视觉-语言-动作基础模型, 计划、推理、组合性, 机械解释性, 知识+大型语言模型, 真实性、忠实性、可信度, 自然语言处理, 人工智能, 具身人工智能
Multimodal Foundation Models, Large Language Models, Vision-Language Foundation Models, Vision-Language-Action Foundation Models, Planning, Reasoning, Compositionality, Mechanistic Interpretability, Knowledge + LLMs, Factuality, Faithfulness, Trustworthiness, NLP, AI, Embodied AI 控制与建模：先进的基于模型/无模型的控制策略和多体动力学。基于物理学的仿真：利用计算机图形学、计算机视觉等基于物理原理的高保真仿真。人工智能与机器学习：在机器人领域进行强化学习、模仿学习、表征学习。
Control & Modeling: Advanced model-based/model-free control strategies and multibody dynamics. Physics-Based Simulation: High-fidelity simulations grounded in physical principles using computer graphics, computer vision etc. AI & Machine Learning: Reinforcement learning, imitation learning, representation learning in robotics. 机械一体化、传感与驱动机器人学、动力学、建模与控制、人机交互、机器学习应用于机器人、将物理学与机器学习相结合以增强机器人平台的控制质量和可靠性，减小物理系统与机器学习之间的差距
Integrated mechatronics, sensor and actuator robotics, dynamics, modeling and control, human-machine interaction, machine learning applied to robotics, blending physics with machine learning to enhance control quality and reliability of robot platforms, reducing the gap between physical systems and machine learning. 鲁棒机器学习，可解释机器学习，数据密集型科学机器学习，基于拓扑结构的鲁棒优化，测试时间适应性，基于原理的优化，可解释机器学习，多模态学习，缺失模态学习，视觉语言基础模型，扩散模型
Robust Machine Learning, Explainable Machine Learning, Data-intensive Scientific Machine Learning, Topology-informed Robust Optimization, Test-time Adaptation, Rationale-informed Optimization, Interpretable Machine Learning, Multimodal Learning, Missing Modality Learning, Vision-Language Foundation Models, Diffusion Models 计算机视觉（例如，注册，分割，分类，检测，合成，成像），多模态数据分析的机器学习（例如，图像，文本，组学），值得信赖和稳健的医疗人工智能（例如，可解释性，泛化性），数据高效的机器学习（例如，高效标注，半监督学习）
Computer Vision (e.g., registration, segmentation, classification, detection, synthesis, imaging), Machine Learning for Multi-Modal Data Analysis (e.g., image, text, omics), Trustworthy and Robust AI for Healthcare (e.g., explainability, generalizability), Data Efficient Machine Learning (e.g., efficient annotation, semi-supervised learning) 荧光材料合成、光学性质表征、分析和机理研究、荧光成像、金属有机框架（MOFs）、共价有机框架（COFs）、有机合成、LC-MS/MS仪器分析、全氟/多氟烷基化合物（PFAS）相关研究。
Fluorescent materials synthesis, optical properties characterization, analysis and mechanism studies, fluorescent imaging, Metal-Organic Frameworks (MOFs), Covalent Organic Frameworks (COFs), organic synthesis, LC-MS/MS instrumentation analysis, per- and polyfluoroalkyl substances (PFAS) related research. 1. 解决随机决策和优化问题中的认知不确定性，通常源自训练数据不足。 2. 应用随机效用函数理论研究市场中消费者的购买偏好。研究集中在提升决策优化和机器学习的鲁棒性。
1. Addressing cognitive uncertainty in stochastic decision-making and optimization problems, typically arising from insufficient training data. 2. Applying stochastic utility function theory to study consumer purchasing preferences in the market. Research is focused on enhancing decision optimization and machine learning robustness. 医学成像（Covid胸透，皮肤癌，乳腺癌，前列腺癌），机器人视觉，小型高效ML模型和训练，嵌入式设备，饮食分析/食品计算，运动分析，遥感，机器人学，SLAM，异常检测，基础设施监测，制造机器健康，强化学习，交通路口分析，昆虫分类，显著性目标分割，视频目标分割
Medical imaging (Covid chest x-ray, skin cancer, breast cancer, prostate cancer), Robotic vision, Small and efficient ML models and training, Embedded device, Dietary analysis/food computing, Sports analytics, Remote sensing, Robotics, SLAM, Anomaly Detection, Infrastructure Monitoring, Manufacturing Machine Health, Reinforcement Learning, Traffic intersection analysis, Insects classification, Saliency object segmentation, Video object segmentation 动力学、控制、优化、人工智能、数据驱动建模、大规模网络动态系统、不确定性、空中交通控制/管理、无人机交通管理、自主空中/地面车辆系统、计算最优控制、随机优化、深度学习、分布式控制/计算、安全自主系统
Dynamics, Control, Optimization, Artificial Intelligence, Data-driven modeling, Large-scale networked dynamical systems, Uncertainty, Air traffic control/management, UAS traffic management, Autonomous air/ground vehicle systems, Computational optimal control, Stochastic optimization, Deep learning, Distributed control/computing, Safe Autonomous Systems 人工智能，空间组学，生物医学成像，癌症诊疗学，计算生物学，生物信息学，生物医学数据科学，生物医学工程，计算机科学，电气工程，统计学，数学，物理学，机器学习，生物医学成像，计算机视觉，单细胞分析技术，组织透明化，光片显微镜，免疫学，癌症系统生物学。
Artificial Intelligence, Spatial Omics, Biomedical Imaging, Cancer Theragnostics, Computational Biology, Bioinformatics, Biomedical Data Science, Biomedical Engineering, Computer Science, Electrical Engineering, Statistics, Mathematics, Physics, Machine Learning, Biomedical Imaging, Computer Vision, Single-cell Profiling Technologies, Tissue Clearing, Light-sheet Microscope, Immunology, Cancer Systems Biology. 1. 骨改建，牙槽骨重塑，牙颌畸形的修复与矫治相关基础研究。2. 颅颌面组织器官缺损修复与再生相关基础研究。3. 口腔疾病相关临床转化研究及应用。4. 同时欢迎交叉方向/背景的同学加入给实验室带来新的思想和技术。
1. Bone remodeling, alveolar bone reshaping, restoration and correction of maxillofacial deformities. 2. Basic research on repair and regeneration of craniomaxillofacial tissue organ defects. 3. Clinical translational research and applications related to oral diseases. 4. Cross-disciplinary students with new ideas and techniques are welcomed to join the lab. - 交通/物流问题的基于学习的模型和算法
- The coupling between population dynamics and infrastructure systems - Transportation electrification and the impacts on community and business resilience - Structure and functional dynamics of transportation and human mobility networks - Efficient and equitable multi-modal mobility systems (with real-world pilots) - Learning-based models and algorithms for transportation/logistic problems 1) 生物工程方法对抗慢性疾病，如肥胖和其他代谢性疾病。2) 包括靶向药物递送和检测皮肤疾病、癌症等生物标志物的先进疗法技术。3) 开发数字分子分析技术，以提高诊断实践。4) 用于可穿戴或可植入生物医学应用的柔性器件。5) 纳米材料及其在生物传感、成像、疗法、催化等方面的应用。
1) Bioengineering approaches to combat chronic diseases such as obesity and other metabolic diseases. 2) Advanced theranostic technologies including targeted drug delivery and detection of biomarkers for skin disorders, cancers, etc. 3) Developing digital molecular analytical technologies for improved diagnosis practices. 4) Flexible devices for wearable or implantable biomedical applications. 5) Nanomaterials and their applications in biosensing, imaging, therapy, catalysis. • Human-centered AI and AI for Social Good. • Trustworthy AI, XAI, and Generative AI. • Multimodal Large Language Model (MLLM), Misinformation Detection and Explanation, Recommender Systems. • Crowdsourcing and Crowdsensing, Human-AI Collective Intelligence. • Federated Learning, Edge Computing, Internet of Things (IoT). • Privacy, Robustness, and FATE (Fairness, Accountability, Transparency, Ethics).
研究方向包括：1) 气候变化和极端天气下的电力系统可靠运行与恢复，2) 数字孪生，大语言模型，与人工智能在电力系统仿真运行的应用，3) 高比例可再生资源渗透下的配电系统运行优化控制。
Research areas include: 1) Power system operation and restoration under climate change and extreme weather conditions, 2) Applications of digital twins, large language models, and artificial intelligence in power system simulation and operation, 3) Optimization control of distribution system operation under high penetration of renewable resources. 机器人学习、具身智能、视觉图形学、（多模态）机器人基础模型、可微分仿真、从仿真到现实以及从现实到仿真、3D视觉、图形学、常识与物理推理、强化学习、终身学习、机械设计、软体机器人、遥操作
Robot Learning, Embodied AI, Vision & Graphics, (Multimodal) Foundation models for Robotics, Differentiable Simulation, Sim2Real and Real2Sim, 3D Vision, Graphics, Commonsense and Physical Reasoning, Reinforcement Learning, Lifelong Learning, Mechanical Design, Soft Robotics, Teleoperation 1. 高效的机器学习算法, 2. 用于AI加速的算法-系统协同设计（包括新兴的深度学习模型，如GNNs，LLMs，Diffusion 模型等），3. 用于芯片设计的大规模机器学习, 4. 高效的隐私保护机器学习
1. Efficient machine learning algorithm, 2. Algorithm-system co-design for AI acceleration (including emerging deep learning models, such as GNNs, LLMs, Diffusion models, etc.), 3. Large scale machine learning for chip design, 4. Energy efficient privacy preserving machine learning 机器学习：生成模型、概率模型、小样本学习、异常检测、OOD分析、自监督学习、图神经网络、大语言模型（LLM）等。数据科学：时间序列、数据流、图、自然语言，及相关的多模态模型。AI应用（AI for Science/Social Good）：生物医学、神经科学、智能医疗、信息物理系统、智能运维等。
Machine Learning: Generative Models, Probabilistic Models, Few-shot Learning, Anomaly Detection, OOD Analysis, Self-supervised Learning, Graph Neural Networks, Large Language Models (LLM). Data Science: Time Series, Data Streams, Graphs, Natural Language, and related multimodal models. AI Applications (AI for Science/Social Good): Biomedical Sciences, Neuroscience, Intelligent Healthcare, Information Physical Systems, Intelligent Operations, etc. 1) 深度学习算法开发与应用，整合图像数据和组学数据，推动数据驱动的科学发现。2) 基于先进的LLM模型（如ChatGPT、LLaMA等）的大型语言模型算法开发与建模，提升对多模态医学数据的理解与应用。
1) Deep Learning algorithm development and application integrating image and omics data, driving data-driven scientific discoveries. 2) Large language model algorithm development and modeling based on advanced LLM models (such as ChatGPT, LLaMA, etc.) for enhancing understanding and application of multimodal medical data. 机器学习，数据挖掘，大语言模型。最近也在扩展一些人机交互和以人为本的人工智能（Human-Centered AI）的方向。 Machine learning, data mining, and their applications to foundation models (e.g., large language models, large vision models, large multimodal models), trustworthy AI (e.g., explainability, fairness, robustness, domain generalization, causal inference, uncertainty quantification, human-AI collaboration), computational medicine (e.g., predictive modeling, medical imaging, clinical NLP, real-world evidence, drug discovery & development).
他的研究领域包括医疗数据挖掘，多模态学习，人工智能安全，LLM和MLLM。研究主题包括LLM增强的多模态医疗数据挖掘和预训练，联邦学习，鲁棒的多模态大语言模型（MLLM）和自动化机器学习。
He works on healthcare data mining, multimodal learning, AI security, LLM, and MLLM. Research topics include LLM-enhanced multimodal healthcare data mining and pretraining, Federated Learning, Robust multimodal large language models (MLLM), and Automated machine learning. 开发新颖的光学成像技术以及心血管研究的计算方法，特别强调心脏结构和功能。跨学科平台包括基于光片/光场成像的采集、基于压缩感知的图像恢复和基于机器学习的图像解释。
Development of novel optical imaging techniques along with computational methods for cardiovascular study, with special emphasis on cardiac structure and function. Interdisciplinary platform includes light-sheet / light-field imaging-based acquisition, compressed-sensing based image recovery, and machine learning-based image interpretation. LLM Agent, AI for social good, NLP for mental health, Question answering, conversational systems, knowledge-grounded NLP
LLM Agent, AI for social good, NLP for mental health, Question answering, conversational systems, knowledge-grounded NLP 研究集中于部署技术，使日常环境更具包容性和共情力，利用人体感知、人工智能、机器人、扩展现实和数字孪生。研究项目包括心理生理监测、可信赖机器人、扩展现实应用、数字孪生和强化学习。
The research focus is on deploying technologies that make daily surroundings more inclusive and empathetic, utilizing human sensing, AI, robots, extended reality, and digital twins. Research projects include psychophysiological monitoring, trustable robots, extended reality applications, digital twins, and reinforcement learning. AI与网络，生成式人工智能与网络优化，强化学习及其在计算机网络系统中的应用，AI支持的语义通信，以用户为中心的人工智能用于网络设计，网络与人工智能，大型人工智能模型的分布式训练与推断，下一代无线传输技术，边缘智能
AI for Network, Generative artificial intelligence and network optimization, Reinforcement learning and its applications in computer networking systems, AI-enabled semantic communications, Human-centric AI for user-centric network design, Network for AI, Distributed training and inference of large AI models, Next-generation wireless transmission technologies, Edge intelligence 人工智能、机器学习、数据挖掘、数据流、在线算法、实时决策系统、图形、拓扑数据挖掘、AI应用于科学领域（材料、分子、城市等）、不确定性量化、因果/贝叶斯推断、基于推理的系统、面向软件工程和编程语言的数据驱动模型
Artificial Intelligence, Machine Learning, Data Mining, Data streams, online algorithms, real-time decision-making systems, Graphs, topological data mining, AI4Science (materials, molecules, cities, etc.), Uncertainty quantification, causal/Bayesian inference, reasoning-based systems, Data-driven models for software engineering and programming languages My research focuses on the societal aspects of machine learning and algorithmic decision-making. I aim to understand the societal implications of machine learning and develop learning systems that are aligned with social norms and reliable in dynamic environments. The research topics of my recent works are: (1) Trustworthy machine learning (e.g., fairness, privacy, security, robustness, interpretability); (2) Learning in uncertain and dynamic environment (e.g., strategic classification, out-of-distribution generalization); (3) Learning from distributed agents (e.g., federated learning)
My focus is on security and privacy issues, with the ultimate goal of developing generalized AI in system/software security to identify vulnerabilities, much like how Noam Brown created the first Poker AI. While I am not limited to blockchain, I find it particularly engaging due to its vast amount of open code and bytecode, transparent data from millions of users and bots, and real-world attacks involving billions of USD. Don't get tricked by the term "blockchain". My past research in blockchain security aligns closely with general system/software security, despite the specialized name.
课题组主要研究半导体器件和集成光子学系统，旨在为新一代成像、通信和计算系统提供更快、更小、更节能的光学器件。具体研究领域包括：1. 新型光源器件（分布式、超快速和可重构激光阵列）；2. 元光学（光学领域的信号和图像处理）；3. 量子光学传感和成像（超越经典极限的传感和成像）。
The research group focuses on semiconductor devices and integrated photonics systems, aiming to provide faster, smaller, more energy-efficient optical devices for the next generation of imaging, communication, and computing systems. Specific research areas include: 1. Novel light source devices (distributed, ultra-fast, and reconfigurable laser arrays); 2. Meta-optics (signal and image processing in the field of optics); 3. Quantum optical sensing and imaging (sensing and imaging beyond classical limits). 分布式缓存系统，持久性缓存，基于CXL和内存池的缓存，NoSQL数据库（RocksDB、LevelDB、HBase），存储系统，文件系统，云存储，新型存储介质（NVM、SMR、DNA、Glass），用于AI/ML服务的数据基础设施。
Distributed caching systems, persistent cache, cache based on CXL and memory pools, NoSQL databases (RocksDB, LevelDB, HBase), storage systems, file systems, cloud storage, new storage media (NVM, SMR, DNA, Glass), data infrastructure for AI/ML services. 研究方向包括深度学习系统和编译器的设计和优化，并行和分布式编程，深度学习应用在异构计算平台上的性能优化，指标驱动的内核优化，深度学习模型在大规模计算平台上的高效映射/调度，容错深度学习系统。
Research areas include deep learning systems and compiler design and optimization, parallel and distributed programming, performance optimization of deep learning applications (e.g., graph neural networks and recommendation systems) on heterogeneous computing platforms (e.g., GPUs), metrics-driven kernel optimizations, efficient mapping/scheduling of deep learning models on large-scale computing platforms, failure-tolerated deep learning systems. Mobile Computing: Wireless Sensing Systems, Human-Computer Interaction Applications, Smart Home and Internet of Things Applications. Cybersecurity: Mobile Authentication, Novel Biometrics, Smart Home and Internet of Things Security.
1. 非凸优化的全局最优解：继续探究一些简单非凸问题的全局自由解。主要研究有没有很好的办法去找到这个最优解，设计一些新的方法和理论，并且用数学语言精准的描述。 2. 新的优化算法：基于我们组已有的一些理论基础，设计并测试一些基于新理论的优化方法（如零阶算法，混合算法），使其可以应付愈加复杂的机器学习模型，比如LLM。 3. 机器学习新理论的研究：探索性研究，探究机器学习应该关注的新属性，比如参数化理论，信噪比理论，并构建一些基础定理。 4. 大模型和新模型: 探究这些复杂模型的optimization landscape，training/finetuning dynamics，以达到更好的训练，finetune它们的效果。同时我们也对基于理论的大模型的简化与优化感兴趣。
1.驾驶行为建模与学习。我们致力于通过AI学习理解微观驾驶行为，并且结合AI和控制方法去优化驾驶行为。 2.智能交通系统中的网络安全。随着自动驾驶车辆的出现，网络安全风险也随之增加。我们的工作包括分析网络攻击对车辆和交通流的研究。 3.现代交通分析中的AI方法。我们的工作不仅限于微观研究，还涉及数据分析，探索更宏观的交通问题。 4.交通流理论研究。 5.交通安全。
可变形结构背后的力学原理，包括折纸启发式结构，软体机器人结构，自适应超材料，多功能机构，可变形结构数值仿真，基于机器学习的逆设计，以及可变形结构的制造和实验方法。
Mechanics behind deformable structures, including origami-inspired structures, soft robotics, adaptive metamaterials, multifunctional mechanisms, numerical simulation of deformable structures, machine learning-based inverse design, and manufacturing and experimental methods for deformable structures. 高效深度神经网络、高性能大模型算法、高性能自监督训练、高性能多模态学习、深度神经网络安全、多智能体强化学习、算法硬件协同设计、深度神经网络训练和推理加速、大模型微调加速、AI算法在EDA上的应用
Efficient deep neural networks, high-performance large model algorithms, high-performance self-supervised training, high-performance multimodal learning, deep neural network security, multi-agent reinforcement learning, algorithm-hardware co-design for deep neural network training and inference acceleration, large model fine-tuning acceleration, AI algorithm applications in EDA 1. 理论计算机科学：近似算法、组合优化、博弈论等 2. 智能科学与技术：深度学习、机器学习、数据挖掘、计算机视觉、自然语言处理等 3. 前沿交叉研究：AI for Science，即计算机科学、智能科学与数学、物理、化学、生物等自然科学的交叉研究
1. Theoretical Computer Science: Approximation Algorithms, Combinatorial Optimization, Game Theory, etc. 2. Intelligent Science and Technology: Deep Learning, Machine Learning, Data Mining, Computer Vision, Natural Language Processing, etc. 3. Frontier Interdisciplinary Research: AI for Science, which involves interdisciplinary research between Computer Science, Intelligent Science, and Mathematics, Physics, Chemistry, Biology, etc. 无线通信，无线通信中的人工智能，空间、空气和地面集成网络，软件定义无线电，网络，5G和Next G通信中的网络安全，无人机群体的空中通信，医疗保健，CV2X，网络物理系统，物联网医疗，可穿戴设备，连接健康的跨学科研究
Wireless Communication, AI in wireless communications, Space, Air and Ground Integrated Networks, Software Defined Radio, Networks, Cybersecurity in 5G and Next G communications, UAV Swarm based Aerial Communications, Interdisciplinary research in healthcare, CV2X, Cyber-physical systems, IoMT, Wearables, Connected Health 计算流体动力学, 非牛顿流体的流变性能模拟和实验, 多相流, 热传导建模, 基础设施对自然灾害的流固耦合模拟, 多功能材料和可持续建筑材料制造工艺, 基础设施材料和韧性中的多目标优化和机器学习
Computational fluid dynamics, rheology of non-Newtonian fluids, multi-phase flow, heat transfer modeling, Fluid-structure interaction modeling in infrastructure resilience to natural hazards, Multi-functional materials and sustainable construction materials and manufacturing process, Multi-objective optimization and machine learning applications in infrastructure materials and resilience 可再生能源整合及电力系统运行优化, 网络化微电网/分布式逆变电源协调控制, 信息物理能源系统建模、仿真、分析与数字孪生, 安全强化学习及多智能体系统, 数据驱动的电力市场设计/决策、需求响应和电动汽车能源管理
Renewable energy integration and power system operation optimization, Networked microgrids/distributed inverter coordination control, Cyber-physical energy system modeling, simulation, analysis, and digital twins, Secure reinforcement learning and multi-agent systems, Data-driven power market design/decisions, demand response, and electric vehicle energy management 2. Exoskeleton 3. Human motion capture and simulation 4. Reinforcement learning and Computer vision 5. Human-robot collaboration 深度学习研究，对话人工智能，语音技术，序列处理，自监督学习，持续学习，协作学习，可解释性神经网络，大规模和小体积模型，多模态基础模型，图神经网络，深度生成模型，神经网络的替代训练方法，分布外泛化
Deep Learning Research for Conversational AI, speech technologies, sequence processing, self-supervised learning, continual learning, cooperative learning, interpretable neural networks, large scale and small-footprint models, multimodal foundation models, graph neural networks, deep generative models, alternative training methods for neural networks, out-of-distribution generalization 材料设计、先进制造、器件制造、水能源纽带周围的空间、生物医学领域、自适应仿生材料合成、纳米材料原位断裂测试、能源、水处理和生物医学器件的光学、机械和电化学性能
Materials design, advanced manufacturing, device fabrication, space around the water-energy nexus, biomedical area, self-adaptive biomimetic materials synthesis, in-situ fracture test of nanomaterials, optical, mechanical, and electrochemical performance of energy, water treatment, biomedical devices 1. Flexible material additive manufacturing 2. Biomimetic flexible materials and structures 3. Functional flexible devices, soft robots 计算社会科学, 人工智能/机器学习, 生成式人工智能, 内容工程, 自然语言处理, 负责任的人工智能, 企业社会责任, 环境社会治理, 信息技术对劳动力与生产率的影响, 人工智能公平性, 市场设计与机制, 网络效应, 平台与生态系统, 偏好学习与推荐系统, 声誉系统与在线评论, 社交网络, 专利与创新, 医疗信息系统, 金融科技
Computational social science, Artificial Intelligence/ML, Generative AI, Content Engineering, Natural Language Processing, Responsible AI, Corporate social responsibility, ESG, Effects of IT on labor and productivity, AI Fairness, Market and mechanism design, Network effects, Platforms and ecosystems, Preference learning and recommender systems, Reputation systems and online reviews, Social networks, Patents and Innovations, Healthcare information systems, Fintech 开发用于生物医学应用的新型光学成像仪器，包括光学相干层析技术，荧光薄层光学断层扫描技术，单光子和双光子显微镜，多模态成像和内窥镜。应用领域包括脑功能成像，肿瘤诊断与治疗监测，组织工程，与医生合作开发和产业化新型医疗仪器。
Development of novel optical imaging instruments for biomedical applications, including Optical Coherence Tomography, Fluorescence Laminar Optical Tomography, single-photon and two-photon microscopy, multimodal imaging, and endoscopy. Applications include brain functional imaging, tumor diagnosis and treatment monitoring, tissue engineering, and collaboration with doctors for the development and industrialization of new medical instruments. 可穿戴机器人设备设计和控制，人体神经肌肉系统建模和控制，基于先进机器学习/加强学习的非线性动力系统控制方法开发，可穿戴辅助或康复机器人设备的线性、非线性、自适应和优化控制算法开发，人机交互系统控制与多传感器融合和多动力学系统耦合
Wearable robotic devices design and control, Human neuromuscular system modeling and control, Nonlinear dynamic control methods based on advanced machine learning/reinforcement learning, Development of linear, nonlinear, adaptive, and optimization control algorithms for wearable assistive or rehabilitation robotic devices, Human-machine interaction system control with multi-sensor fusion and multi-dynamic system coupling 车载自组织网络(VANETs) - 机器学习用于智能自动驾驶车辆，边缘计算和物联网的网络安全，基于无人机的地下情况感知的无线传感器网络，联网和自动驾驶车辆的网络安全，智能认知无线电网络及其安全性
Vehicular Ad Hoc Networks (VANETs)-enabled Machine Learning for Intelligent Autonomous Vehicles, Cyber Security for Edge Computing and IoTs, Drone-based Wireless Sensor Network for Underground Situation Awareness, Cyber Security for Connected and Autonomous Vehicles, Intelligent Cognitive Radio based Networks & Its Security 实时调度、强化学习（RL）、数字孪生应用于自动驾驶系统（例如，ROS2）、深度学习模型压缩、时序信号处理与预测应用于辅助医疗系统、形式化方法用于实时+智能系统的可靠性、能效性和安全性
Real-time scheduling, Reinforcement Learning (RL), Digital Twin applied to Autonomous Driving Systems (e.g., ROS2), Deep Learning Model Compression, Time Series Signal Processing and Prediction applied to Assistive Medical Systems, Formal Methods for Real-time + Intelligent Systems reliability, energy efficiency, and security