Reinforcement Learning (RL) is considered to be the holy grail on the road to Artificial General Intelligence. It is a field of AI research with the goal of creating agents that learn to act intelligently when applied to environments such as video games. These agents need to plan ahead and take smart actions in order to maximize the gathered rewards. During the Motius Discovery Camp, we applied the latest RL algorithms such as "Mu Zero" by DeepMind to several use cases ranging from Flappy Bird to robotic simulations. The exciting results are presented in this recording from the Motius Discovery Conference in March 2021.
Do you want to build the products of the future?