Reinforcement learning common use cases, recommendation engine, productivity - Susan Shu Chang the data scientist show#039
The Data Scientist Show - Daliana Liu
00:00
How to Find a Right Reward and Ere?
Reinforcement learning is a way of teaching people how to behave. But it can be very difficult to find the right reward for this kind of behavior. Ritvik: We could build in something that deducts the amount of, you know, the company will spend on this action that the reinforcement loarning agent is recommending. And also the lifetime value, or like the present value, of, like, the customer. It takes into account that if you give them a lot of money and they just curn right away, then that's really bad for the reimborsement learning agents.
Transcript
Play full episode
