A new algorithm for machine learning and artificial intelligence
报告专家：夏志宏教授 （美国西北大学）
报告时间：10月21日（星期一）上午10：00-11：00
报告地点：数学学院西303报告厅
报告摘要：We propose a novel machine learning algorithm inspired by complex analysis. Our algorithm has a better mathematical formulation and can approximate universal functions much more efficiently. The algorithm can be implemented in two self-learning neural networks: The CauchyNet and the XNet. The CauchyNet is very efficient for low-dimensional problems such as extrapolation, imputation, numerical solutions of PDEs and ODEs. The XNet, on the other hand, works for large dimensional problems such as image and voice recognition, transformer and large language models. We implemented our algorithm for many scenarios, showing that it is very efficient and acurate. It is much better than than many popular PINN (Physically Inspired Neural; Network) models in various scientific computations; It outperforms KAN (Kolmogorov Arnold Network); For a set of medical image we tested, it can increase accuracy from 88% to 98%. Our algorithm is currently being tested on large language models. Small scale testing shows great promise.
专家简介：夏志宏教授，美国西北大学Pancoe讲席教授，大湾区大学（筹）讲席教授，国际知名数学家和天文学家。主要研究动力系统，天体力学，曾解决百年数学难题Painlevé猜测，天体运动的混沌性，Hamilton系统的通有性质和遍历理论中关于拓扑熵的一系列问题。获得多项重大学术奖励，其中包括美国总统青年研究者奖，Sloan Research Fellowship，首届Blumenthal纯数学进步奖等奖项。2015年创立南方科技大学数学系，以及参与建立未来科学大奖。
邀请人：王宝富
