报 告 人：林宙辰 教授，北京大学

报告时间：7月4日（周日）14:30

报告地点：校本部东区计算机学院大楼1106

邀 请 人：马丽艳 副研究员 

报告摘要：

The design of convolutional neural networks (CNNs) has undergone two phases: manual design at the early stage, which requires much engineering insights, and the automatic search at the current stage, which heavily relies on computing power. Whether there is an underlying theory for designing good CNNs becomes a crucial research problem. In this talk, I will illustrate our efforts on pursuing this goal. Although I haven’t found a unified principle that can result in all the effective CNNs, I do find multiple principles that can help design CNNs from various aspects. 

报告人简介:

林宙辰，北京大学教授，IAPR/IEEE Fellow，国家杰青，中国图象图形学学会机器视觉专委会主任，中国自动化学会模式识别与机器智能专委会副主任。研究领域为机器学习、计算机视觉和数值优化。发表论文200余篇，英文专著2本。多次担任CVPR 、ICCV、NIPS/NeurIPS、ICML、IJCAI、AAAI和ICLR领域主席，任ICPR 2022共同程序主席，曾任IEEE T. PAMI编委，现任IJCV、Optimization Methods and Software编委。