【帮好朋友发帖】NYU-强化学习/多智能体/智能交通

avatar 808298
pixel555
988
2
帮我在NYU的好朋友发帖!本人超有趣, Twitter上看看他的帖子就知道(@EugeneVinitsky)

研究方向 (包括但不限于)
- 自动化验证 (how can we create closed-loop validation systems that continually find and fix agent failures)
- 真实、安全的智能体 (Building human-like agent models to improve our simulation and safety capabilities)
- 大规模多智能体(Scaling up MARL with GPU-based simulators.)

Preview


Preview


导师介绍
尤金·维尼茨基是纽约大学城市与土木工程学院的教授,苹果公司的研究科学家,同时也是C2SMARTER联盟的成员。他的主要研究兴趣是通过开发新的学习算法、软件和工具,尽可能简化多智能体控制器、规划者和智能的开发过程。他寻找这些技术在土木工程问题和自主性方面的应用。他从加州大学伯克利分校获得了控制工程博士学位,从加州理工学院获得了物理学士学位,并曾在苹果公司和Facebook AI Research担任研究科学家,同时还曾在DeepMind和特斯拉自动驾驶部门实习。他以前的工作包括:

  • 使用强化学习设计控制器,为在田纳西州高速公路上部署的100辆部分自动化车辆服务(circles-consortium.github.io
  • 许多流行的模拟器,如FLOW(flow-project.github.io)和Nocturne(github.com),一种自动驾驶模拟器
  • MAPPO论文证明了PPO在许多多智能体问题上是最先进的技术


联系方式
NYU的Transportation Engineering or Civil Engineering Department 申请截止时间是12月1号
投递邮箱:1point3acres.com
邮件主题请注明:[Prospective Student] PhD Application from {名字}。
请附上你的CV、成绩单,可以附上一小段你的自我介绍以及研究兴趣 【报我名字说不定有加成!】

Less formal bio 【不那么正式的自我介绍,英文是原版】

Hi, I'm Eugene. My core belief about the world is that simple, easy-to-scale methods of automation are the best way to build a better future. Many of the problems that are exciting now are single-agent problems but I focus on multi-agent problems because they are futuristic and exciting; they present ways to build new systems far behind what humans are currently capable of.
【大噶好,我是尤金,我想做简单高效的方法解决真实世界的多智能体问题】

I started as a physics PhD but fell in love with controls and RL and switched to doing that at UC Berkeley. Now I'm at NYU where we're trying to figure out how to use RL and learning methods to improve the efficiency and safety of our built infrastructure and transportation systems.
【俺是伯克利毕业的,现在在NYU】

I chose to go to academia over industry because I love being part of a close-knit team and I hope to create that feeling within my group.
【我最后选择了学术界因为我喜欢紧紧团结在一起的感觉,我在NYU的组也会是这样的氛围!】

In terms of students that fit very well with me, I like people who have strong learning towards empirical work over theoretical work and love to program and build things.
【欢迎学习能力强,动手能力强/理论能力不错的同学来勾搭我!】

Finally, in my spare time, I read a ton of sci-fi, play every sport I possibly can, or play the guitar. Looking forward to meeting some of you!
【我平常读超多科幻小说,喜欢运动,弹吉他,希望你也和我一样有趣!】
2条回复