Intro to Machine Learning
What are two reasons why we want to incorporate learning into an
agent-based system? - ANS>1) Agent to solve task without exactly
knowing all the scenarios that it will find itself in
2) Agent that can solve task but in the future are wanting to have it
make predictions of the system (Stock market)
What are the three types of learning? - ANS>1) Unsupervised Learning:
Learns patterns in the input even though no explicit feedback is
supplied
2) Reinforcement Learning: The agent learns from a series of
reinforcements - rewards or punishments
3) Supervised Learning: Agent observes some example input-output
pairs and learns a function that maps from input to output.
Explain the difference between training data and testing data. -
ANS>Training data is a subsect of the entire data set that is used to train
an agent to find a hypothesis function that is then tested on a testing
data set, which is the remainder of the data set.
