AlphaZero is a computer program that was originally built to master the games of chess, shogi and go. It was trained solely via "self-play", in parallel, with no access to opening books or endgame tables.

The Amazon.com e-commerce system identifies users of similar tastes in terms of their profiles and shopping orders. In this way, the marketing analysts understand the specific interests of each group of users and customize the ads personalized exclusively to those users.

Apple iPhone’s FaceID application matches a query face against a database of registered faces, typically employed to authenticate users through ID verification services. In this way, legitimate users can unlock a device, make payments, access sensitive data, and provide detailed facial expression tracking for Animoji and other features

AlphaZero is reinforcement learning, right? I think because AlphaZero improved and became increasingly stronger and better at learning and decision-making

The Amazon is unsupervised learning, right?

Apple iPhone’s FaceID application is supervised learning, right?


I think you are correct. Here is a simple way to distinguish between the types of learnings:

  1. Is the learning using some training data or not? If it is, then it is most likely supervised learning. If it isn't, there are two options
  2. Assume the answer to the last question was false. Is there an interaction with the "environment" when the agent learns? If there is, it is reinforcement learning. If there isn't, then this is most likely unsupervised learning.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.