Description:
California Partners for Advanced Transportation Technology (PATH) is a research center in the Institute of Transportation Studies at University of California, Berkeley, and has been a leader in Intelligent Transportation Systems (ITS) research since its founding in 1986. PATH is a multi-disciplinary program with staff, faculty, and students from universities worldwide and cooperative projects with private industry, state and local agencies, and nonprofit institutions. In conjunction with Berkeley DeepDrive, we continue to push the scientific forefronts of automated driving systems (ADS), robotics, computer vision (CV), and artificial intelligence (AI) and machine learning (ML). We are seeking qualified and passionate candidates to fill one or more positions of postdoctoral researchers, to integrate Deep Learning (DL) into ADS studies.
Appointment:
This is a 12-month, 100%-time appointment, with possible extension based on project needs. Anticipated start date: October 2019 or sooner, pending on candidate availability
Responsibilities:
The postdoctoral researcher will work closely with the PIs to explore the application of Deep Learning methods, e.g. Reinforcement Learning (RL), Inverse Reinforcement Learning (IRL), Adversarial Learning, Meta Learning, Semi-supervised Learning, etc., on Autonomous Driving studies such as decision-making, trajectory planning, driving control, and driving styles learning, for scenarios on highways and urban streets. The candidates may participate in future projects on intelligent autonomy beyond autonomous driving.
The main duties and responsibilities will consist of:
(1) Methodology Formulation
Design cutting-edge and practical algorithms based on solid understanding of state-of-the-art DL algorithms to address AD related problems. Particularly, combine the merits of RL, IRL, GAN, etc., to learn driving policies for autonomous vehicles.
(2) Code Implementation
Implement the formulated algorithms, develop simulation tools, and conduct data analysis with popular computer programming languages (e.g., Python, C/C++) and deep learning development frameworks (e.g. TensorFlow, Pytorch).
(3) Autonomous Vehicle Model Design
Understand and build vehicle kinematic & dynamic models and control & safety constraints for vehicle related problem optimization.
(4) Preparing proposals for project funding and reports for project review.
(5) Writing articles for refereed academic journals and/or conferences.
Basic Qualifications Required (At the time of application):
Candidates must have completed all degree requirements except the dissertation or be enrolled in an accredited Ph.D or equivalent international degree program.
Additional Required Qualifications (By start date):
Ph.D or equivalent international degree.
A Record of peer reviewed publications.
Preferred Qualifications (By start date):
(1) Ph.D in Computer Science, Mechanical Engineering, Electrical Engineering, or related fields.
(2) Research experience in DL based projects, including deep reinforcement learning, inverse reinforcement learning, imitation learning, and generative adversarial learning, with a portfolio demonstrating past work experience and deliverables.
(3) End-to-end experience with most aspects of research (problem formulation, system requirements, data generation, data analysis, verification, reporting).
(4) Record of peer reviewed publications relevant with deep learning and/or autonomous driving.
(5) Excellent writing and communication skills.
(6) Proposal writing experiences and skills.
Salary:
Commensurate with experience, and based on UC Berkeley salary scales. This position provides full postdoctoral scholar benefits.
Application deadline:
Job will remain open until filled, but applicants are encouraged to submit by 8/31/2019.
Application Procedure:
To apply, send email to pin_wang@berkeley.edu.
Job Location:
Richmond and Berkeley, CA
Document Requirements:
Cover Letter
Curriculum Vitae - Your most recently updated CV
Statement of Research Experience
Contacts:
Please direct questions to: pin_wang@berkeley.edu.
All letters will be treated as confidential per University of California Policy and California State law.