返回信息流标题:Quantile Interpolation of Distributions
讲者:Prof. Alex Pang, University of California, Santa Cruz
时间:2014年11月27日(星期四)下午3:00-4:00
地点:北京大学理科二号楼2736
内容简介:
Both Monte Carlo simulations and multiple observations, measurements, or expe
riments produce what's called an ensemble data set. Ensemble data represents
both the data and uncertainty about the data. They are gaining popularity a
nd are now routinely used in weather forecasting for example. Analyzing such
data sets, particularly, with the proper handling of the uncertainty aspect
still requires further research. This talk will focus on the aspect of inter
polating pointwise distributions derived from ensemble data sets. We will di
scuss the desirable properties of interpolation and look at a number of alter
native interpolation methodss. Finally, we will present an efficient bivaria
te interpolation technique based on interpolating quantile curves of CDF surf
aces.
讲者介绍:
Alex Pang is a Professor of Computer Science at UC Santa Cruz. He received hi
s PhD in Computer Science from UCLA in 1990, and his BS in Industrial Enginee
ring from University of the Philippines with magna cum laude in 1981. His res
earch interests are in comparative and uncertainty visualization, flow and te
nsor visualization, and collaborative visualization. His research has been su
pported by various funding agencies such as NSF, ONR, Darpa, DOE, LANL, and N
ASA, as well as industrial partners such as Sun and HP.
Professor Pang has received a certificate of recognition for previous NASA wo
rk, as well as an excellence in teaching award from UCSC. He served as an ass
ociate editor of the IEEE Transactions on Visualization and Computer Graphics
, papers co-chair for IEEE Visualization 2006 and 2007, and UCSC Chief Scient
ist for CITRIS from 2006-2007.
报名参加者请邮件至pkuvis@pku.edu.cn,提供姓名和单位信息,以便安排会务
这是一条镜像帖。来源:北邮人论坛 / paper / #18052同步于 2014/11/26
该镜像源已超过 30 天没有更新,可能在源站已被删除。
Paper机器人发帖
[公告]【可视化讲座】Quantile Interpolation of Distributions
luminous
2014/11/26镜像同步0 回复
订阅后,新回复会通过你的通知中心匿名送达。
0 条回复
暂无回复 · 你可以订阅本帖等待新回复。