Want to learn what makes autonomous robots tick? Then you might enjoy hearing more about an area of AI called reinforcement learning.
I haven't worked much with reinforcement learning yet, so I've been on the lookout for a good lecture series on the subject. This one recently caught my attention, because most of the lectures are taught by folks from UC Berkeley and DeepMind.
UC Berkeley is in the top 4 or 5 schools in the US with a graduate program in artificial intelligence.
DeepMind is a UK group of AI researchers that Google acquired in 2014, and they've done some really brilliant work in the field. You might remember them for creating an AI that defeated the world's top Go champions at their own game. They're also known for drawing their inspiration from biological systems. For example, they used fMRI to study how humans navigate their way through a virtual subway network. They concluded that the brain splits the task of completing a journey into different jobs, and different parts of the brain handle different parts of the problem. The cortex got more involved when the participant had to change lines. Other parts of the brain got involved when the participant got closer to the final destination. You can read more about that interesting study in Neuron: http://www.cell.com/neuron/fulltext/S0896-6273(16)30057-5.
The lectures in the online bootcamp were given August 2017, so they will hopefully cover some state-of-the-art research in reinforcement learning.
The first video in this lecture series can be found below. Enjoy!
Credit: Deep RL Bootcamp
