Computers and Technology, 31.03.2021 01:00, hannah2718

We would like to use a Q-learning agent for Pacman (as you did in PA3), but the state space for a large grid is too massive to hold in memory. To solve this, we will switch to feature-based representation of the Pacman game state (similar to PA3 Q10), where we assume that Q(s, a) can be expressed as a (weighted) linear combination of state-action features: Q(s, a)= Σωifi(s, a)

Required:
Suppose we design two features to represent the state (independent of actions): f (s) is the number of ghosts within one step of Pacman, and fp(s) is the number of food pellets within one step of Pacman. Note that we do not have any features dependent on a. Why might this be a bad idea?

Answers: 3

Show answers

Other questions on the subject: Computers and Technology

Computers and Technology, 21.06.2019 14:20, ashl3yisbored

The concept of comes from the acknowledgment that data changes form and often gets copied, moved, and stored in many places. sensitive data often leaves the protection of application databases and ends up in e-mails, spreadsheets, and personal workstation files.

Answers: 3

continue

Computers and Technology, 22.06.2019 15:00, taylorsamodell3217

Who is the first president to use social media as part of his campaign strategy

Answers: 1

continue

Computers and Technology, 23.06.2019 07:00, bskyeb14579

Why is investing in a mutual fund less risky than investing in a particular company's stock? a. mutual funds only invest in blue-chip stocks. b. investments in mutual funds are more liquid. c. mutual funds hold a diversified portfolio of stocks. d. investments in mutual funds offer a higher rate of return.

Answers: 2

continue

Computers and Technology, 24.06.2019 16:00, bsrlee1115

Which type of cloud computing offers easily accessible software and applications on the machines

Answers: 1

continue

Do you know the correct answer?

We would like to use a Q-learning agent for Pacman (as you did in PA3), but the state space for a la...