Management of investment portfolios employing reinforcement learning

There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

Abstract

Investors are presented with a multitude of options and markets for pursuing higher returns, a task that often proves complex and challenging. This study examines the effectiveness of reinforcement learning (RL) algorithms in optimizing investment portfolios, comparing their performance with traditional strategies and benchmarking against American and Brazilian indices. Additionally, it was explore the impact of incorporating commodity derivatives into portfolios and the associated transaction costs. The results indicate that the inclusion of derivatives can significantly enhance portfolio performance while reducing volatility, presenting an attractive opportunity for investors. RL techniques also demonstrate superior effectiveness in portfolio optimization, resulting in an average increase of 12% in returns without a commensurate increase in risk. Consequently, this research makes a substantial contribution to the field of finance. It not only sheds light on the application of RL but also provides valuable insights for academia. Furthermore, it challenges conventional notions of market efficiency and modern portfolio theory, offering practical implications. It suggests that data-driven investment management holds the potential to enhance efficiency, mitigate conflicts of interest, and reduce biased decision-making, thereby transforming the landscape of financial investment.

Related collections

Most cited references 47

Record: found
Abstract: found
Article: not found

Long Short-Term Memory

Jürgen Schmidhuber, Jürgen Schmidhuber (2003)

Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient-based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O(1). Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.

0 comments Cited 8312 times – based on 0 reviews      Review now

Record: found
Abstract: found
Article: not found

Human-level control through deep reinforcement learning.

Volodymyr Mnih, Koray Kavukcuoglu, David Silver … (2015)

The theory of reinforcement learning provides a normative account, deeply rooted in psychological and neuroscientific perspectives on animal behaviour, of how agents may optimize their control of an environment. To use reinforcement learning successfully in situations approaching real-world complexity, however, agents are confronted with a difficult task: they must derive efficient representations of the environment from high-dimensional sensory inputs, and use these to generalize past experience to new situations. Remarkably, humans and other animals seem to solve this problem through a harmonious combination of reinforcement learning and hierarchical sensory processing systems, the former evidenced by a wealth of neural data revealing notable parallels between the phasic signals emitted by dopaminergic neurons and temporal difference reinforcement learning algorithms. While reinforcement learning agents have achieved some successes in a variety of domains, their applicability has previously been limited to domains in which useful features can be handcrafted, or to domains with fully observed, low-dimensional state spaces. Here we use recent advances in training deep neural networks to develop a novel artificial agent, termed a deep Q-network, that can learn successful policies directly from high-dimensional sensory inputs using end-to-end reinforcement learning. We tested this agent on the challenging domain of classic Atari 2600 games. We demonstrate that the deep Q-network agent, receiving only the pixels and the game score as inputs, was able to surpass the performance of all previous algorithms and achieve a level comparable to that of a professional human games tester across a set of 49 games, using the same algorithm, network architecture and hyperparameters. This work bridges the divide between high-dimensional sensory inputs and actions, resulting in the first artificial agent that is capable of learning to excel at a diverse array of challenging tasks.

0 comments Cited 1901 times – based on 0 reviews      Review now

Record: found
Abstract: found
Article: not found

Mastering the game of Go without human knowledge

David M. Silver, Julian Schrittwieser, Karen Simonyan … (2017)

A long-standing goal of artificial intelligence is an algorithm that learns, tabula rasa, superhuman proficiency in challenging domains. Recently, AlphaGo became the first program to defeat a world champion in the game of Go. The tree search in AlphaGo evaluated positions and selected moves

0 comments Cited 836 times – based on 0 reviews      Review now

All references

Author and article information

Contributors

Flavio Barboza

Journal

Journal ID (nlm-ta): PeerJ Comput Sci

Journal ID (iso-abbrev): PeerJ Comput Sci

Journal ID (publisher-id): peerj-cs

Title: PeerJ Computer Science

Publisher: PeerJ Inc. (San Diego, USA )

ISSN (Electronic): 2376-5992

Publication date (Electronic): 11 December 2023

Publication date Collection: 2023

Volume: 9

Electronic Location Identifier: e1695

Affiliations

[1 ]Faculdade de Engenharia Elétrica, Universidade Federal de Uberlândia , Uberlândia, MG, Brazil

[2 ]Faculdade de Gestão e Negócios, Universidade Federal de Uberlândia , Uberlândia, MG, Brazil

Article

Publisher ID: cs-1695

DOI: 10.7717/peerj-cs.1695

PMC ID: 10773882

PubMed ID: 38192465

SO-VID: db95ed1e-678d-49a8-8e99-d7072c284066

License:

This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. For attribution, the original author(s), title, publication source (PeerJ Computer Science) and either DOI or URL of the article must be cited.

History

Date received : 7 August 2023

Date accepted : 23 October 2023

Funding

Funded by: Sapiens Agro (Sapiens Inteligência Ltda)

The authors received no funding for this work. Sapiens Agro (Sapiens Inteligência Ltda) provided funding for the APC. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Management of investment portfolios employing reinforcement learning

Read this article at

Abstract

Related collections

Computer Vision, Deep Learning, Deep Reinforcement Learning, IoT

Most cited references 47

Long Short-Term Memory

Human-level control through deep reinforcement learning.

Mastering the game of Go without human knowledge

Author and article information

Contributors

Journal

Affiliations

Article

History

Funding

Categories

Comments

Comment on this article

Similar content 263

Most referenced authors 527