Postdoc Position for Deep Reinforcement Learning Flow Control

Job Description

Position: Postdoctoral Fellow

Topic: Reinforcement Learning for Vortical Flow Control and Sensing. 

Requirements:

  1) Age under 35.

  2) Ph.D. degree (obtained or about to obtain) in the fields of Applied 

Mathematics/Fluid Mechanics/Computer Science/Robotics and Control, or other 

related fields.

  3) Passion and commitment towards scientific research; Good communication skills 

and ability to work independently and effectively in a highly collaborative 

research environment; Ability to independently conduct research and supervise 

graduate students.

  4) Good English-speaking skills and writing skills.

  5) Background in Fluid-structure interaction, and Reinforcement learning is 

favored

Compensation and Benefits

  The research team offers a competitive compensation package commensurate with 

the selected candidate’s qualifications and experience. Applications for relevant 

projects and programs will be encouraged and supported.

  Postdoc(s) selected for China Postdoctoral Science Foundation or Zhejiang 

Province Selected Funding for Postdoctoral Research Projects will be entitled to 

further receive a supporting fund of the same amount from the Hangzhou municipal 

government.

  Postdoc(s) who work in Hangzhou on a full-time basis after completing their 

postdoctoral research will be eligible for applying an allowance of 400,000RMB 

from the Hangzhou municipal government.

How to Apply

  To apply, please send your CV and a brief research statement (less than 2 pages) 

in English in PDF format to fandixia@westlake.edu, and indicate “Postdoc 

Application” in the email subject.

Introduction of the Lab/ Research Field

  At i4-FSI Lab (Intelligent, Informational, Integrative, Interdisciplinary Fluid-

Structure Interaction Laboratory), we focus on the fundamental understanding of 

fluid-structure interaction phenomena, bio-inspired amphibious robots design as 

well as AI application in vortical flow control and sensing. Based on physics-

informed (and -informative) machine learning, we combine domain expertise (fluid 

mechanics, robotics, and control) and proper machine learning tools to address the 

inherent spatial and temporal non-linearity and multiscality of fluid-related 

problems at a greater scale and a broader scope. 

  We envision a research paradigm shift in fluid mechanics to a physics-informed 

(and -informative) probabilistic learning framework, which leads to disruptive 

technology transformation in the aerospace and marine industry to a more 

efficient, safe, and eco-friendly future. Some of the current research topics 

focus on advancement of reinforcement learning algorithm for vortical flow control 

and sensing.