Q&A: What are the main resources to learn AI?
Welcome to this week’s Beautifeye Q&A. We sat down once more with Luca, our CEO and AI expert, to ask a few pressing questions that we received in recent days.
The first question of today’s episode is: what are the main resources to learn AI?
Luca says “I think one hidden gem for AI research enthusiasts is the website videolectures.net. So, videolectures is an amazing collection of technical videos from talks given at conferences all around the world.
Otherwise, the series of Andrew NG classes on Coursera is a very great place to start.
The 2nd question of today is: why is beating the Go world champion a significant milestone for AI?
You might have heard it, recently Google beat the world champion of the ancient Chinese game called “Go”. I think this is extraordinary under several points of view. First of all, Go is a very complicated game, even for humans. Secondly, I think it’s impossible to solve it with a brute force approach for a very simple reason: there are more possible moves in Go than there are atoms in the universe.
Thirdly, Alpha Go didn’t only beat the world champion once, but it also beat the six top world players, showing incredible supremacy over humans. Finally, while in chess, for instance, you can win by following quite simple rules, in Go, choosing a good move demands a great deal of intuition. And as you know, intuition is very difficult to reproduce with a machine. So, well done Google!
The third question of today: what are the ethical issues with using AI?
Well, I would say many and most parts of them are largely unsolved. Actually, I think the debate hasn’t started yet. Take for instance autonomous driving: if behind the wheel there’s an algorithm, who’s responsible? If there’s an accident, who is at fault? Is it the owner of the car, the manufacturer or the algorithm itself? So, these are big problems that can significantly undermine and delay the adoption of Artificial Intelligence.
That concludes this week’s interview with our CEO. We’ll be back next week with more answers on AI, Machine Learning, Computer Vision and Deep Learning, so keep your suggestions coming!