基本信息
浏览量:29
职业迁徙
个人简介
I was a PhD student in the School of Computer Science at Carnegie Mellon. Along with my advisor Drew Bagnell, I worked on extending boosting and ensemble methods to a variety of machine learning applications with an eye towards applications in robotics and vision. As of 2014, I am now working at Google Pittsburgh.
My main research project and thesis work was focused on the anytime prediction problem. Building off of traditional boosting techniques, we designed prediction algorithms which are capable of efficiently using any test-time computation budget to generate predictions that are near-optimal with respect to that budget. Unlike other approaches which trade cost for accuracy and vice-versa, our anytime approach generates a single predictor which can dynamically fit any budget. Furthermore, this budget need not be known apriori at training time.
Previously I was advised by Dave Touretzky. Together we worked on planning, mapping and navigation for educational robotics using Tekkotsu, the software framework that he and his students have developed.
研究兴趣
论文共 5 篇作者统计合作学者相似作者
按年份排序按引用量排序主题筛选期刊级别筛选合作者筛选合作机构筛选
时间
引用量
主题
期刊级别
合作者
合作机构
UAI'16: Proceedings of the Thirty-Second Conference on Uncertainty in Artificial Intelligence (2016): 279-288
引用3浏览0EI引用
3
0
Alexander Grubb, Drew Bagnell
引用101浏览0EI引用
101
0
作者统计
合作学者
合作机构
D-Core
- 合作者
- 学生
- 导师
数据免责声明
页面数据均来自互联网公开来源、合作出版商和通过AI技术自动分析结果,我们不对页面数据的有效性、准确性、正确性、可靠性、完整性和及时性做出任何承诺和保证。若有疑问,可以通过电子邮件方式联系我们:report@aminer.cn