Developer
Tesla reinforcement learning agent
Tabular Q-learning with discretised states for buy, hold and sell decisions.
Overview
A learning project applying tabular Q-learning to trading actions.
The question
Explore decisions whose outcomes depend on a sequence of actions.
My approach
Represent trading as discretised states and buy, hold or sell actions.
Outcome and limits
Exploration of sequential trading decisions.
A learning project; no investment performance claim or financial recommendation.
Interested in this work? Get in touch.