Machine Learning for Self-Driving Cars: Behavioral Cloning (1/2)

Where: Foo Cafe, sign-up via

When: Thu 20 Apr 2017 at 17.45 (Malmö/Stockholm time)

Updated description:

Deep Learning (Deep Neural Networks) is a family of techniques within Machine Learning and Artificial Intelligence that is widely used to teach cars to drive autonomously. For instance, Deep Learning is used for object detection i.e. identifying pedestrians, vehicles, traffic signs, and lane lines. At this event, we are teaching a car to drive itself by training on data recorded by a human driver. The only input available to the car to choose the steering angle is a photo of the road ahead. No hints are given about useful intermediate signs (lane lines, road edges etc.); only the steering angle of a human driver in similar situations. Mimicking human driving in this way is called Behavioural Cloning and has proven to be a very powerful tool in the development of self-driving cars.

This event can be considered as part 1 and will be continued on the 9th of May, a bit more technical / mathematical - check it out!

Event outline: 17:45 to 18:00 - Meet & Greet 18:05 - Short introduction of the meetup

18:10 - Machine Learning for Self-Driving Cars

18:55 - Pizza & Beer

19:20 - Demo 19:40 - Networking & Mingle

20:30 - over & out


218 Words

2017-04-20 17:45 +0200