293: True Personalization Through Reinforcement Learning
Super Data Science: ML & AI Podcast with Jon Krohn
00:00
The Advantages and Disadvantages of Reinforcement Learning
Reinforcement learning doesn't require this supervised data, it is much more biased resistant. Sometimes we come across a problem that an artificial intelligence might be exhibiting sexism or racism. In my view, like I'd love to get your opinion, this reinforcement learning learns on its own and therefore it wouldn't be able to pick up that human bias from the supervised data. That's very true. And reinforcement learning model can also get stuck in an environment and only learn like a very, very limited kind of like behavior. But you can basically easily fix that by expanding the world or expanding the environment that your agent is going to see.
Play episode from 32:33
Transcript
