Is There a Value Bias in Machine Learning?

The whole process of machine learning is you're trying to minimize this error function, right, in the output. And so one way to set that up for investing is, is to give it fundamental data for each company at each point in time. And then as the output, did it outperform the market over the subsequent year or not? Then you feed that to the machine learning algorithms and it tries to minimize the error of predicting whether a company would have outperformed the market if they'd been given more information. It's just what the data told them works over the long run.

Transcript

Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app