Sign Up For Free Webinars

Applying Deep-learning to Solving Autonomous Driving

Tyrone Wilkinson
NYC Data Science Academy Fellow
Panelist Spotlight

Tyrone Wilkinson is a recent NYC Data Science Academy graduate who intends to impact the world with AI positively and ultimately contribute to the advent of AGI. While he has a bachelor’s in Computer Science from Columbia University, his career in IT had mainly been uneventful until he decided it was time to pivot directions in life via the Bootcamp. Now, he looks forward to tackling complex problems like autonomous driving and solving them for the benefit of humankind.

Inspired by the problem of autonomous driving, Tyrone focused on one aspect of it: human attentiveness, which is required at all times when operating sub-level 4 autonomous vehicles. He gathered, prepared, and labeled 10K+ images and fine-tuned the top layers of a pre-trained ResNet-152 model to classify attentiveness in images, videos accurately, and live feed. He also produced an app that allows anyone to get their videos labeled and scored with an "Attention Score."Β  Learn what Tyrone has learned at the Academy and how his project has a host of real-world applications outside of improving driver attentiveness.Β 
  • Employers can implement attention monitoring to increase workplace productivity.Β 
  • Video chat platforms like Zoom can incorporate it as a feature.Β 
  • Students can use it to fine-tune their workflow.Β 
There are many possibilities.
Explore More
Student Presentation

NYC Data Science Academy Alumnus, Andrew Dodd will give a free introductory workshop to learn how to structure a machine learning (ML) problem and build it into a server-client model with AWS ...

Continue 55 mins
Student Presentation

an overview of Kubernetes, and give a walkthrough on how to start deploying ML applications at scale. Kubernetes is a container orchestration platform for managing workloads, ...

Continue 63 mins
Student Presentation

Insights on how to aggregate and analyze mortgage data geographically using the Freddie Mac single family loan level dataset, HUD Crosswalk files and Census Bureau data.

Continue 76 mins