A4.3 Machine learning approachesA4.3.6
How an agent learns to make decisions by interacting with its environment in reinforcement learning
Describe how an agent learns to make decisions by interacting with its environment in reinforcement learning
AHL≈ 55 minCommand term: DescribePremium
In this lesson
- Describing the agent–environment interaction in terms of states, actions, rewards and policies
- Describing the principle of cumulative reward (the return)
- Describing the exploration versus exploitation trade-off and strategies for balancing it
This lesson is part of Premium
Every topic has one free lesson. Premium unlocks every interactive lesson, the exam-style questions at the end of each one, and the rest of Praxis IB.
What's inside
- LearnThe agent–environment loop
- LearnCumulative reward
- LabStep through the loop
- CheckQuick check
- LearnExploration versus exploitation
- LabExplore or exploit?
- CheckExploring or exploiting?
- LearnAgents in the real world
- ExamWorked exam answer
- ReviewMistakes examiners see
- Exam practiceExam-style questions
Ends with 4 exam-style questions worth 13 marks, with a self-marking markscheme.