Reinforcement learning (RL) gained world fame as a powerful machine learning solution to problems deemed, until very recently, too complex to be solved by computers. However, RL has yet to show that it can be as transformational in real applications - outside of stylized examples (e.g., the world of board or video games). In this talk, Pedro will present his work in developing a reinforcement learning application at easyJet and describe his lessons, including the challenges and the huge potential of this approach.
Pedro Alves, Data Scientist at EasyJet:
Pedro is a data scientist at easyJet where he applies Machine learning and statistics to business problems on several areas from revenue management to network optimization. He holds a PhD in Economics from the London School of Economics where he developed optimal pricing algorithms based on game theory and statistical learning techniques.