Skip to content

3.2 RL Environment

Marc Juchli edited this page Apr 22, 2018 · 25 revisions

“(Approximately) Markovian nature of trade execution: if our state space is properly defined, the optimal action at any given point in time is (approximately) independent of any previous actions.” [Kearns et. al.]

E.g. executions do not affect the market for future executions.

RL Overview

The environment, defined as ctc-executioner-v0, is a child class of gym.Env where the functions step and reset are implemented such that we can simulate execution behaviour.

Observation

State

Agent

Action Space

Action

Policy

Clone this wiki locally