人工智能学院2021年系列学术活动——美国康涅狄格大学梁冠男博士学术报告

报告题目：非凸稀疏学习问题的有效算法

Efficient Algorithms for Nonconvex Sparse Learning Problems

报告简介：Sparse learning plays important roles in the fields of statistical learning, machine learning (ML) and signal processing. Learning with high dimensional data often relies on the sparsity-driven regularization. Solving a sparsity-regularized empirical risk minimization (ERM) problem can derive a ML model with sparse parameters, meaning that the model parameter vector has many zero entries. Thus, it helps select relevant features for use in a ML model. For example, via the sparsity regularization, genomic analysis can identify a small number of genes contributing to the risk of a disease, and smartphone-based healthcare systems can detect the most important mobile health indicators. In this report, sparse learning is formulated as nonconvex and/or nonsmooth optimization problems depending on the specific regularizer. We develop efficient stochastic gradient descent (SGD) based algorithms to solve these problems.

稀疏学习在统计学习、机器学习和信号处理等领域发挥着重要作用。高维数据的学习通常依赖于产生稀疏的正则化。求解稀疏正则化约束的经验风险最小化(ERM)问题可以得到参数稀疏的机器学习模型，即模型参数有许多零项。因此，它有助于筛选机器学习模型中相关特征。例如，通过稀疏正则化，基因组分析可以识别一小部分导致疾病风险的基因，基于智能手机的医疗保健系统可以检测到最重要的健康指标。在本报告中，稀疏学习被定义为非凸和/或非光滑数学优化问题。针对这些问题，我们提出了高效的基于随机梯度下降(SGD)的算法。

报告人简介：Guannan Liang obtained his Ph.D. degree at the Computer Science and Engineering  Department of the University of Connecticut in 2021. He received his M.S. degree in Statistics at the University of California, Davis in 2016, and his B.S. degree in Mathematics at Zhengzhou University (China) in 2013. He has published several papers in top-tier machine learning and data mining conference -- Neurips, AAAI, ICDM and CIKM. His primary Ph.D. research focuses on mathematical optimization, scalable machine learning.

梁冠男， 2021年在美国康涅狄格大学取得计算机科学与工程系的博士学位。2016年获美国加州大学戴维斯分校统计学硕士学位，2013年获郑州大学数学学士学位。在顶级机器学习和数据挖掘会议——Neurips, AAAI, ICDM和CIKM上发表了多篇论文。他的主要博士研究方向是随机优化、机器学习。

报告时间：2021年4月16日（星期五）上午9:10-10:10

报告地点：水务集团大厦403室

主办单位：人工智能学院