NVIDIA: Tech Talk

  • Gates Hillman Centers
  • ASA Conference Room 6115
  • Senior Research and Development Engineer
Career Presentation

Machine Learning for Path Perception in Autonomous Driving

Explore the future of AI computing in an engaging talk on "Machine Learning for Path Perception in Autonomous Driving."

The NVIDIA New Jersey team focuses on developing deep neural networks and other machine learning models for path perception in autonomous driving tasks. Using only annotated recordings of human drivers we can train these models to learn important maneuvers, such as lane keeping, lanes changes, and handling highway on/off-ramps. Furthermore, we can show that these networks have learned to react strongly to consistent and reasonable features, such as lane markers and other vehicles, without any semantic segmentation labels. We will also look at simulation tools we develop for validating our trained models and measuring their performance over time in a realistic manner that reflects their real-world driving performance.

Alperen Degirmenci is a Senior Research and Development Engineer at NVIDIA, working on developing machine learning models and tools for autonomous driving. He earned his Ph.D. and M.S. degrees in Electrical Engineering from Harvard University in 2018 and 2015 with a secondary field in Computational Science and Engineering, and his B.S. degree in Mechanical Engineering from the Johns Hopkins University in 2012, with minors in mathematics, computer science, robotics, and computer-integrated surgery. Before joining NVIDIA, Alperen worked at the Harvard Biorobotics Lab on real-time, high-performance algorithm development for medical ultrasound image processing and robotic procedure guidance in catheter-based cardiac interventions.

We will be raffling off an NVIDIA GeForce RTX 2080 SUPER!

5:45 pm - 6:00 pm: Check-in and enter raffle
6:00 pm – 6:40 pm: Tech Talk
6:40 pm – 7:00 pm:  Q&A + NVIDIA GeForce RTX 2080 SUPER Raffle

RSVP/Registration Requested

For More Information, Please Contact: