AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Is There a Sweet Spot for Machine Learning?
The idea is that we don't actually advise people to use macine learning to giveout the base trailing tgy, because it's a very easy to overfit. We ask them to use this for plice management and capital evocation. So in owords, everybody will have their own cres of wor. And you know, i can give you my craspor won't do you any good, because i'm not taing yours y. The process of feature selection will automatically eliminate thought fundamentalsif they find out that you are trading a fore mor, it will not be useful. I wil automatically eliminate. It's easier to narrow it down than to build it up.