牛菓，女，副教授，碩士生導(dǎo)師, 博士碩士均就讀于中山大學(xué)，本科畢業(yè)于解放軍信息工程大學(xué)。實(shí)驗(yàn)室關(guān)注人工智能學(xué)習(xí)研究，研究聚焦于大模型，多模態(tài)，圖像生成，視頻理解，同時(shí)包括流形學(xué)習(xí)、對(duì)抗學(xué)習(xí)、目標(biāo)檢測(cè)、圖像分類等。
近年來(lái)，主持的國(guó)家自然基金項(xiàng)目1項(xiàng)，省級(jí)項(xiàng)目2 項(xiàng)，佛山市博士項(xiàng)目1項(xiàng)目。參與省重點(diǎn)、省市級(jí)項(xiàng)目5項(xiàng)。參與多項(xiàng)省重點(diǎn)科研計(jì)劃。以第一作者在國(guó)內(nèi)外期刊與會(huì)議發(fā)表高水平學(xué)術(shù)論文13篇，申請(qǐng)國(guó)家發(fā)明專利3項(xiàng)（其中1項(xiàng)已授權(quán)）。實(shí)驗(yàn)室配備多臺(tái)高性能GPU服務(wù)器，滿足深度學(xué)習(xí)與大模型訓(xùn)練需求，鼓勵(lì)學(xué)生參與課題研究、發(fā)表論文、撰寫專利，支持科研競(jìng)賽與學(xué)術(shù)交流。團(tuán)隊(duì)協(xié)作與交流氛圍輕松積極，定期組織組會(huì)，與中山大學(xué)、騰訊公司科研團(tuán)隊(duì)合作密切，同時(shí)注重學(xué)生科研能力與綜合素質(zhì)的共同發(fā)展。
目前組內(nèi)研究方向多元化，課題靈活，組內(nèi)研究氛圍輕松，并充分鼓勵(lì)討論與合作，歡迎對(duì)人工智能學(xué)習(xí)研究感興趣的同學(xué)加入！

[1]主持 2020 年國(guó)家自然科學(xué)基金項(xiàng)目-青年項(xiàng)目，結(jié)題，2021.01-2023.12。
[2]主持2022年廣東省粵佛聯(lián)合基金項(xiàng)目-區(qū)域項(xiàng)目，在研，2022.10-2025.10。
[3]主持2023年廣東教育廳-特色創(chuàng)新項(xiàng)目，在研，2023.10-2025.10
[4] 主持2019年廣東省粵佛聯(lián)合基金項(xiàng)目-青年項(xiàng)目，結(jié)題，2020.01-2022.12。

[1]Guo Niu, Zhengming Ma,and Shuyu Liu, A Multikernel-Like Learning Algorithm Based on Data Probability Distribution, Mathematical Problems in Engineerin, 2016 (SCI).
[2] Guo Niu, Zhengming Ma,and Shaogao lv, Ensemble multiple-kernel based manifold regularization, Neural Processing Letters, 2017 (SCI).
[3] Guo Niu, Zhengming, Ma, Local non-linear alignment for non-linear dimensionality reduction, IET Computer Vision,?2017. (SCI)
[4] Guo Niu , Meikang Qiu, and Yanchun Gu, Kernel Learning Method on Riemannian Manifold with Geodesic Distance Preservation, the 4th IEEE International Conference on Big Data Security on Cloud, 2018.
[5] Guo Niu and Zhikui Duan, Inner Product Optimization for Effective Multiple KernelLearning, IOP Conf. Series: Earth and Environmental Science (2019).
[6] Guo Niu, Yanchun Gu, and Zhikui Duan, Semi-Supervised Learning Based on Local Adaptive Kernels, 2019 IEEE Smart World Congress (SmartWorld 2019), Leicester, United Kingdom, 2019
[7] Guo Niu?and?Zhengming Ma, Tensor Dimensionality Reduction via Mode Product and HSIC, IET Image Processing, 2021,?(SCI)DOI: 10.1049/ipr2.12285
[8] Guo Niu?and?Zhengming Ma, Local Quasi-Linear Embedding Based on Kronecker Product Expansion of Vectors, Journal of Intelligent & Fuzzy Systems ,?2021?.
[9] Guo Niu, ?Zhengming Ma, HaoQing Chen, and Xu, Su, Polynomial approximation to manifold learning, Journal of Intelligent and Fuzzy Systems,DOI:10.3233/JIFS-200202, 2021
[10] Guo Niu?and?Zhengming Ma, Tensor local linear embedding with global subspace projection optimisation, IET Computer Vision, DOI:10.1049/cvi2.12083, 2022.
[11] Guo Niu?and?Zhengming Ma, Local homeomorphism regularized non-negative Tucker decomposition of tensor data, Journal of Intelligent and Fuzzy Systems, DOI:10.3233/JIFS-220785, 2022
[12] Guo Niu, Nannan Zhu, Xueqing Fang, Xueyue Lei，Holographic Radar Tracking Algorithm for ‘Low-Slow-Small’ Maneuvering Targets，IET International Radar Conference 2023.
[13] Guo Niu, Nannan Zhu, Zhengming Ma, Xin Wang, Xi Liu, Yan Zhou, Yuexia Zhou, RKHS reconstruction based on manifold learning for high-dimensional data, Applied Intelligence, vol. 55, issue 124, 2025
[14]牛菓; 王修才; 顧艷春; 段志奎; 陳建文; 樊耘; 一種基于測(cè)地距離的黎曼流形保持核學(xué)習(xí)方法及裝置, 2021-8-31, 中國(guó), ZL2018 1 0542508.2 (已授權(quán))
[15]牛菓，邵強(qiáng)，周月霞，顧艷春；基于流形結(jié)構(gòu)核函數(shù)的多核學(xué)習(xí)方法、系統(tǒng)及相關(guān)設(shè)備，202211048789.9，
[16] 牛菓，周月霞，邵強(qiáng)，樊耘，陳建文；一種基于流形結(jié)構(gòu)核函數(shù)的方差最大化子空間學(xué)習(xí)算法，202311143716.2
[17]牛菓、焦帥威（研究生）、周燕、周月霞、何志敏、易長(zhǎng)安；一種圖像分類識(shí)別方法、系統(tǒng)、終端設(shè)備和存儲(chǔ)介質(zhì),202411928459.8
[18] 牛菓(7/7),佛山高新技術(shù)進(jìn)步獎(jiǎng)二等獎(jiǎng)。

[email protected]