Sebastian Thrun is one of the greatest roboticists, computer scientists, and educators of our time. He led development of the autonomous vehicles at Stanford that won the 2005 DARPA Grand Challenge and placed second in the 2007 DARPA Urban Challenge. He then led the Google self-driving car program which launched the self-driving revolution. He taught the popular Stanford course on Artificial Intelligence in 2011 which was one of the first MOOCs. That experience led him to co-found Udacity, an online education platform. He is also the CEO of Kitty Hawk, a company working on building flying cars or more technically eVTOLS which stands for electric vertical take-off and landing aircraft. ~ Lex Fridman's Website
Note 1: Building An Intelligence System
Sebastian Thrun mentioned that machine learning is the needed ingredient to build an intelligent machine. The previous version of Artificial Intelligence was built using Expert Systems but this requires that the rules and actions be explicit, which might not be the case. For instance, what are the rules for walking? With machine learning, the system can now learn to pick up the rules and regularities by themselves.
Note 2: DARPA Challenge
In preparation for the DARPA Challenge, Sebastian Thrun's team was a victim of a bug and it took them ages to find it. This lead to the team, setting up a testing team. This testing team is responsible for testing the self-driving car, to eliminate the bugs immediately when surfaced by the tests. The testing team came up with 160 tests for the self-driving vehicle to go through. Sebastian Thrun also mentioned they focused very much on improving the weakness so that they have an overall robust system.
Note 3: Good Leadership
Over here, Sebastian Thrun shared what is good leadership through his DARPA challenge and his current work experience. He mentioned there is a difference between humans and machines. Humans have pride, aspirations, emotions, etc. Good leaders are those that connect to the people they work with, make them feel great about their work. Good leaders turn what their followers desire in life into productive action, and empower their team.
Note 4: Turning Beginners into Experts
In a particular section, there was a discussion on the potential impact Machine Learning can bring to humankind. Previously in Note 1, it was mentioned that machine learning can through large amounts of data picks up patterns. In other words, Machine Learning can "observe" our current experts, pick up both implicit and explicit rules and patterns from expert's behavior and judgment. This trained behavior and judgment may take the expert a long time to gain. Malcolm Gladwell mentioned in his book, "Outliers", to become an expert need 10,000 hours. This translates to non-stop practice and learning for 417 days or 1 year and 1.5 months.
The potential is after machine learning is used to pick up the implicit and explicit rules and patterns, they can then be passed to professionals who just joined the field. This can elevate their expertise, giving them the learnings, skills and experience of the experts before them.
Sebastian Thrun is a very humble person and he is very genuine in wanting to use Artificial Intelligence for the betterment of the human race. One thing that I totally concur with him and that is education should, as much as possible, be available to everyone. Personally I see education as a tool for humans to advance, for people to get out of poverty and move up the social ladder. I see education as a tool for many to build a better life for themselves and their families.
The interview was rather short and I was hoping he can touch on more about the technical details of self-driving cars, what are the possible challenges when self-driving cars going to rely more on computer vision to drive rather than using scanners.
If you will like to discuss the topic on AI, feel free to Tweet me or add me on LinkedIn. To keep up to date with my learning and sharing, consider subscribing to my newsletter below. Each subscription is a vote of confidence in my work.