GAMES Webinar 2020 – 145期(可视化专题) | Bei Wang(University of Utah)

GAMES Webinar 2020-145期】(可视化专题)

报告嘉宾：Bei Wang(University of Utah)

报告时间：2020年7月2日上午10:00-11:00（北京时间）

报告题目：TopoAct: Visually Exploring the Shape of Activations in Deep Learning

报告摘要：

Deep neural networks such as GoogLeNet and ResNet have achieved impressive performance in tasks like image classification. To understand how such performance is achieved, we can probe a trained deep neural network by studying neuron activations, that is, combinations of neuron firings, at any layer of the network in response to a particular input. With a large set of input images, we aim to obtain a global view of what neurons detect by studying their activations. We ask the following questions: What is the shape of the space of activations? That is, what is the organizational principle behind neuron activations, and how are the activations related within a layer and across layers? Applying tools from topological data analysis, we present TopoAct, a visual exploration system used to study topological summaries of activation vectors for a single layer as well as the evolution of such summaries across multiple layers. We present visual exploration scenarios using TopoAct that provide valuable insights towards learned representations of an image classifier. This is joint work with Archit Rathore, Nithin Chalapathi, and Sourabh Palande. https://arxiv.org/abs/1912.06332

讲者简介：

Bei Wang is currently an Assistant Professor in the School of Computing and a faculty member in the Scientific Computing and Imaging(SCI) Institute. Her research expertise lies in the theoretical, algorithmic, and application aspects of data analysis and data visualization, with a focus on topological techniques. In particular, her research leverages topological data analysis, which provides a a strong theoretical basis for transforming large, complex data into compact, structure-highlighting representations. Such representations connect naturally with and provide infrastructures for data visualization, and inspire the rethinking of interactive data exploration to facilitate analytical reasoning. Some of her current research activities draw inspirations from topology, geometry, and machine learning, in studying vector fields, tensor fields, high-dimensional point clouds, networks, and multivariate ensembles.

讲者个人主页: http://www.sci.utah.edu/~beiwang/

主持人简介：

秦红星，重庆大学计算机学院教授。2008年博士毕业于上海交通大学，2008年至2009年在美国罗格斯新泽西州立大学CAIP从事博士后研究。研究方向主要是几何处理与可视化，目前课题侧重于点云数据的配准、骨架提取与采样。详情请见个人主页：http://www.escience.cn/people/qinhongxing/index.html

GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播？”及“如何加入GAMES微信群？”的信息；
GAMES主页的“资源分享”有往届的直播讲座的视频及PPT等。
观看直播的链接：http://webinar.games-cn.org