AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
How Can You Learn From Your Own Estimate?
There's been, you know, a series of breakthrough starting in really, the 19 seventies. There's an idea called temporal difference learning, which says, rather than waiting till you actually get the reward, you can learn from your own estimate changing. Increasingly, social media companies like facebook are using reinforcement learning to model how they send no cations out. Ye ad, i guess howas hows reinforcement, letting warkt wat?