8.1 Deep Learning

A formal framework that defines probability using three fundamental rules, ensuring consistency in measuring uncertainty. 🎲

Motivation

Problem

One parameter vector \(\mathbf{w}\) is not enough to represent \(s\).

What happens if \(s\) is an image tensor?

Solution

We would like to use neural networks \(\theta\) for more parameters to improve our function approximation:

Based on your intuition, do you think it makes sense to use raw pixels as input for a Deep Q-Network (DQN) instead of preprocessed state features?

Solution