Reinforcement Learning and
Artificial
Intelligence (RLAI)
Jeopardy quiz #1 assignment
In the spirit of learning, and doing things until we can do them
right from memory, i offer this assignment form of the in-class
jeopardy quiz. Unless you got all of the questions right (and i
don't think any of you did) print out this page and fill in the
appropriate question underneath each answer. If you got a
question exactly right in class then you don't need to do it again here
(but it wouldn't hurt).
This should be quick and easy. I will publish my answers (or
rather questions) on the web page after. -10 points for asking
how much this assignment will be worth toward your grade.
Basic entities of RL
100
The thing that makes decisions in the Reinforcement Learning framework
200
The thing that represents the external world in the Reinforcement Learning framework
100
The signal from the agent to the environment
200
The signal from the environment to the agent, in the general case
400
The thing that represents the agent's knowledge of the environment's dynamics
Theory and ideas
100
The purpose of life, according to the Reinforcement Learning framework
200
An action taken to gain information about the environment, without the expectation that it will high obtain reward
300
The action believed to attain the most total reward in the long term
400
The theory that reward is the ultimate goal, that the policy is a means
for obtaining reward, that the value function is a means for finding
the policy, and that the model is a means for finding the value function
500
The proposal that all goals can be thought of as maximizing the sum of
the future values of an externally received scalar signal.
The field of Reinforcement Learning
100
Marvin Minsky, Ian Witten, Andrew Barto, Chris Watkins, Gerry Tesauro, and perhaps yours truly
200
Psychology, Operations Research, Neuroscience, and Artificial Intelligence
