Machine learning in the wild
Thoughtworks Technology Podcast
00:00
Reinforcement Learning for Dynamic Pricing
The easiest way to do reinforcement learning is to have access to a simulator. You can play against yourself, who can play against a computer and become iteratively better. For real world problems, let's say pricing, you don't have a simulator. If you knew what people would buy at what price, you wouldn't need machine learning to begin with.
Transcript
Play full episode
