时间:2011-09-21 来源:综合办 编辑:qqli 访问次数:2067
报告题目:Machine Learning Theory and Applications
报告人:Dr. Zhiyuan Luo,英国伦敦大学Royal Holloway学院计算机系
报告地点:智能系统与控制研究所(教十八)223室
报告时间:2011年9月21 日(星期三)上午9:00 - 10:40
Abstract:
The lecture series will introduce modern machine learning techniques and
develop the basic theoretical background in learning methods. Two well
known learning methods, k-nearest neighbours and Support Vector Machine
will be discussed in detail. Finally some theoretical and experimental
developments in building computable approximations of Kolmogorov's
algorithmic notion of randomness will be presented. Based on these
approximations a new set of machine learning algorithms have been
developed, for example, conformal predictors. The algorithms allow us to
make predictions and estimate confidence for high-dimensional data when
the only assumption is the i.i.d. assumption: the data are independent and
identically distributed.
欢迎广大师生参加!
报告人简介:骆志远博士,英国Heriot-Watt大学博士,伦敦大学Queen Mary 学院博士后,现任伦敦大学Royal Holloway学院计算机系高级讲师。曾任英国大东电报局网络模型顾问,不仅在学术界,在工业界有着丰富研究开发经历,作为首席科学家承担过英国皇家学会、欧共体、英国兽医实验局环境食品和农村事物部、英国工程和物理科学研究委员会和英国医学研究委员会多个重大科学研究项目,近五年来承担科研项目总经费超过56万英镑(合560人民币万元),发表论文60余篇,2006和2008年两次分别获得电器和电子工程学会颗粒计算国际大会(IEEE International Conference on Granular Computing)和第九届智能数据工程和自动学习国际大会(9th International Conference on Intelligent Data Engineering and Automated Learning)最佳论文奖。