RL Concepts

Question 1

What does the “Step Size” variable “alpha” do ?

Answer 1

Influences the rate of learning

Question 2

Difference between stationary vs non-stationary reward ?

Answer 2

Stationary rewards are constant. Non-stationary rewards vary.

Question 3

What does DeepMind’s AlphaTensor algorithm do ?

Answer 3

Speeds up matrix multiplication via Reinforcement Learning

Question 4

What is Bootstrapping?

Answer 4

Using one or more estimated values of a variable to update estimates of the same variable