Join for free and connect with our local tech scene

Stay on top of the latest companies and upcoming events with our weekly newsletter, and be counted among the people building the future of your local tech community.

Pittsburgh, PA /

Intro to Software Safety for Machine Learners

Code & Supply Community Center 5648 Friendship Ave, 3rd Floor , Pittsburgh, PA 15206 (map)

• What we'll do

C&S will host the Pittsburgh Machine Learners. For the most up to date information, check out their page for this event at https://www.meetup.com/Pittsburgh-Machine-Learners/events/245474789/

Machine learning is not only becoming an essential part of our everyday lives, it is now becoming responsible for keeping us safe. For example, self driving cars will use machine learning to use ensure the safety of their passengers, passengers in other cars on the road, and nearby pedestrians. Over the past few decades, a community of software safety practitioners has learned how to build high integrity, trustworthy software. The problem is, many of these techniques are fundamentally incompatible with machine learning. Today we are seeing the worlds of machine learning and traditional software safety begin to collaborate. As someone working at the intersection of these two worlds, I would like to use this talk as an opportunity to foster more such collaboration in the Pittsburgh machine learning community. To work that end, I will introduce key incompatibilities between techniques used in machine learning and traditional software safety. I will then survey approaches proposed for overcoming these problems.

Michael Wagner is the CEO of Edge Case Research, a company founded to make autonomous systems safer. He has 20 years of experience building autonomous vehicles starting at Carnegie Mellon University. He built lunar rovers for Red Whittaker, autonomous scientists that explored Antarctica, and self driving technology for tackling harsh off road terrain. Ten years ago he and Philip Koopman began researching techniques to design and verify autonomous systems. Today Michael applies this experience at Edge Case Research with the goal of providing validation technology for autonomy across a range of industries including self driving cars, materials handling, and robotic work cells.

Comments

Sign in to comment.