Skip to content
All library documents

Advanced FinRL Tutorials on Reinforcement Learning and Portfolio Allocation

Article FinRL

Summary

This advanced tutorial section is intended for readers already familiar with FinRL or its market simulation companion, or who have worked through introductory notebooks. It points to a comparison of three deep reinforcement learning libraries supported by FinRL, aimed at readers interested in their differing implementations and capabilities.

A second notebook builds an ensemble agent from several popular deep reinforcement learning algorithms and compares it with other agents on a portfolio allocation task. The page also references a tutorial on explainable deep reinforcement learning for portfolio allocation. It is only an index: it summarizes the topics of the notebooks but gives no algorithms, experimental setup, results, or evidence about which agent performs best. Readers must consult the referenced materials to assess methodology and limitations.

Key ideas

  • The section targets users with prior exposure to FinRL or its introductory materials.
  • One notebook compares three deep reinforcement learning libraries supported by the framework.
  • Another constructs an ensemble agent and compares it with other agents on portfolio allocation.
  • The page lists tutorial topics but provides no experimental methods or comparative results.

Tags

Machine learningPortfolio constructionEquities
Full text
# 2 Advance


:github_url: https://github.com/AI4Finance-Foundation/FinRL

2-Advance
========================

This section is recommended for users with some familiarity of FinRL or FinRL-Meta (or already run the notebooks in "1-Introduction").

Notebooks in this section includes:

1. `FinRL_Compare_ElegantRL_RLlib_Stablebaseline3.ipynb <https://github.com/AI4Finance-Foundation/FinRL/blob/master/tutorials/2-Advance/FinRL_Compare_ElegantRL_RLlib_Stablebaseline3.ipynb>`_

In this notebook, we compare the three DRL libraries that supported in FinRL. Users who know these DRL libraries might find this interesting.

2. `FinRL_Ensemble_StockTrading_ICAIF_2020.ipynb <https://github.com/AI4Finance-Foundation/FinRL-Tutorials/blob/master/2-Advance/FinRL_Ensemble_StockTrading_ICAIF_2020.ipynb>`_

In this notebook, we implement an "ensemble agent", which is a ensemble of several popular DRL algorithms. Then we compare the performance of the ensemble agent and other DRL agents on the portfolio allocation task.

`FinRL_PortfolioAllocation_Explainable_DRL.ipynb <https://github.com/AI4Finance-Foundation/FinRL-Tutorials/blob/master/2-Advance/FinRL_PortfolioAllocation_Explainable_DRL.ipynb>`_.

Shown in full with attribution under the source's licence. Licence: MIT

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.