VALSE

VALSE 首页 活动通知 好文作者面授招 查看内容

20180418-9 徐易:Accelerated Stochastic Subgradient Methode under Local Error

2018-4-12 21:09| 发布者: 程一-计算所| 查看: 3553| 评论: 0

摘要: 报告嘉宾:徐易(The University of Iowa)报告时间:2018年04月18日(星期三)早上10:30(北京时间)报告题目:Accelerated Stochastic SubgradientMethode under Local Error Bound Condition主持人:钟燕飞(武汉 ...

报告嘉宾:徐易(The University of Iowa)

报告时间:2018年04月18日(星期三)早上10:30(北京时间)

报告题目:Accelerated Stochastic Subgradient Methode under Local Error Bound Condition

主持人:张利军(南京大学


报告人简介:

爱荷华大学计算机系博士在读,本科毕业于浙江大学统计学专业。

讲者个人主页:

https://homepage.cs.uiowa.edu/~yxu71/


相关文献: 

1. Stochastic convex optimization: faster local growth implies faster global convergence, Yi Xu, Qihang Lin, Tianbao Yang, In ICML, 2017.


报告摘要:

In this talk, I will introduce two accelerated stochastic subgradient methods for stochastic non-strongly convex optimization problems by leveraging a generic local error bound condition. The novelty of the proposed methods lies at smartly leveraging the recent historical solution to tackle the variance in the stochastic subgradient. The key idea of both methods is to iteratively solve the original problem approximately in a local region around a recent historical solution with size of the local region gradually decreasing as the solution approaches the optimal set. The difference of the two methods lies at how to construct the local region. The first method uses an explicit ball constraint and the second method uses an implicit regularization approach. For both methods, the improved iteration complexity in a high probability for achieving an ϵ-optimal solution is established. Besides the improved order of iteration complexity with a high probability, the proposed algorithms also enjoy a logarithmic dependence on the distance of the initial solution to the optimal set. 


特别鸣谢本次Webinar主要组织者:

VOOC责任委员:钟燕飞(武汉大学)

VODB协调理事:张利军(南京大学


活动参与方式:

1、VALSE Webinar活动依托在线直播平台进行,活动时讲者会上传PPT或共享屏幕,听众可以看到Slides,听到讲者的语音,并通过聊天功能与讲者交互;

2、为参加活动,请关注VALSE微信公众号:valse_wechat 或加入VALSE QQ群(目前A、B、C、D、E、F、G群已满,除讲者等嘉宾外,只能申请加入VALSE H群,群号:701662399),直播链接会在报告当天(每周三)在VALSE微信公众号和VALSE QQ群发布;

*注:申请加入VALSE QQ群时需验证姓名、单位和身份,缺一不可。入群后,请实名,姓名身份单位。身份:学校及科研单位人员T;企业研发I;博士D;硕士M。

3、在活动开始前10分钟左右,讲者会开启直播,听众点击直播链接即可参加活动,支持安装Windows系统的电脑、MAC电脑、手机等设备;

4、活动过程中,请勿送花、打赏等,也不要说无关话语,以免影响活动正常进行;

5、活动过程中,如出现听不到或看不到视频等问题,建议退出再重新进入,一般都能解决问题;

6、建议务必在速度较快的网络上参加活动,优先采用有线网络连接;

7、VALSE微信公众号会在每周一推送上一周Webinar报告的总结及视频(经讲者允许后),每周四发布下一周Webinar报告的通知。


最新评论

小黑屋|手机版|Archiver|Vision And Learning SEminar

GMT+8, 2024-4-18 09:19 , Processed in 0.015581 second(s), 15 queries .

Powered by Discuz! X3.4

Copyright © 2001-2020, Tencent Cloud.

返回顶部