Level III · Module 10
Reinforcement Learning for Trading Decisions
Learning to act from rewards: Bellman equations, Q-learning and policy gradients, and their use in execution and market making.
This module is locked
Level III enrollment is open to Level II students, or with the complete three-level program.
Explore the module outline below, or start with the free orientation and Module 1.
Read the free orientation and Module 1Module outline
- Markov decision processes
- Q-learning and policy gradients
- Execution, simulators and evaluation
Intended learning outcome: Formulate a trading decision as an RL problem and judge whether an RL result is credible.