Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading

We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but in...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cornalba, Federico, Disselkamp, Constantin, Scassola, Davide, Helf, Christopher
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title
container_volume
creator Cornalba, Federico
Disselkamp, Constantin
Scassola, Davide
Helf, Christopher
description We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but incorporated in the learning process). Firstly, using several important assets (cryptocurrency pairs BTCUSD, ETHUSDT, XRPUSDT, and stock indexes AAPL, SPY, NIFTY50), we verify the reward generalization property of the proposed Multi-Objective algorithm, and provide preliminary statistical evidence showing increased predictive stability over the corresponding Single-Objective strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge over the corresponding Single-Objective strategy when the reward mechanism is sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss the generalization properties with respect to the discount factor. The entirety of our code is provided in open source format.
doi_str_mv 10.48550/arxiv.2203.04579
format Article
fullrecord <record><control><sourceid>arxiv_GOX</sourceid><recordid>TN_cdi_arxiv_primary_2203_04579</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2203_04579</sourcerecordid><originalsourceid>FETCH-LOGICAL-a679-21820bf73a6aab401ad33b14ac6fd8b29a034ab78c1819946033f1342485f0f03</originalsourceid><addsrcrecordid>eNotkLlOxDAQht1QoIUHoMIvkOArFx1arpWCVkLbR2NnvDJKnMgx4ah5cLKBavQf-qX5CLniLFVllrEbCJ9uToVgMmUqK6pz8vPy3kWX7PUbmuhmpAE_ILT0iB4DdO4bohv8Ld31Yxhm5490xGCH0IM3SAdL7xFH-orOL6bBHn2kNULwp-piURjHzpl1ZaLO02kJOkxgmjDSGKBd9AU5s9BNePl_N-Tw-HDYPif1_mm3vasTyIsqEbwUTNtCQg6gFePQSqm5ApPbttSiAiYV6KI0vORVpXImpeVSieV3yyyTG3L9N7tiaMbgeghfzQlHs-KQv-ZYXWo</addsrcrecordid><sourcetype>Open Access Repository</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype></control><display><type>article</type><title>Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading</title><source>arXiv.org</source><creator>Cornalba, Federico ; Disselkamp, Constantin ; Scassola, Davide ; Helf, Christopher</creator><creatorcontrib>Cornalba, Federico ; Disselkamp, Constantin ; Scassola, Davide ; Helf, Christopher</creatorcontrib><description>We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but incorporated in the learning process). Firstly, using several important assets (cryptocurrency pairs BTCUSD, ETHUSDT, XRPUSDT, and stock indexes AAPL, SPY, NIFTY50), we verify the reward generalization property of the proposed Multi-Objective algorithm, and provide preliminary statistical evidence showing increased predictive stability over the corresponding Single-Objective strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge over the corresponding Single-Objective strategy when the reward mechanism is sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss the generalization properties with respect to the discount factor. The entirety of our code is provided in open source format.</description><identifier>DOI: 10.48550/arxiv.2203.04579</identifier><language>eng</language><subject>Computer Science - Learning ; Quantitative Finance - Computational Finance ; Quantitative Finance - Trading and Microstructure</subject><creationdate>2022-03</creationdate><rights>http://arxiv.org/licenses/nonexclusive-distrib/1.0</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>228,230,776,881</link.rule.ids><linktorsrc>$$Uhttps://arxiv.org/abs/2203.04579$$EView_record_in_Cornell_University$$FView_record_in_$$GCornell_University$$Hfree_for_read</linktorsrc><backlink>$$Uhttps://doi.org/10.48550/arXiv.2203.04579$$DView paper in arXiv$$Hfree_for_read</backlink></links><search><creatorcontrib>Cornalba, Federico</creatorcontrib><creatorcontrib>Disselkamp, Constantin</creatorcontrib><creatorcontrib>Scassola, Davide</creatorcontrib><creatorcontrib>Helf, Christopher</creatorcontrib><title>Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading</title><description>We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but incorporated in the learning process). Firstly, using several important assets (cryptocurrency pairs BTCUSD, ETHUSDT, XRPUSDT, and stock indexes AAPL, SPY, NIFTY50), we verify the reward generalization property of the proposed Multi-Objective algorithm, and provide preliminary statistical evidence showing increased predictive stability over the corresponding Single-Objective strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge over the corresponding Single-Objective strategy when the reward mechanism is sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss the generalization properties with respect to the discount factor. The entirety of our code is provided in open source format.</description><subject>Computer Science - Learning</subject><subject>Quantitative Finance - Computational Finance</subject><subject>Quantitative Finance - Trading and Microstructure</subject><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2022</creationdate><recordtype>article</recordtype><sourceid>GOX</sourceid><recordid>eNotkLlOxDAQht1QoIUHoMIvkOArFx1arpWCVkLbR2NnvDJKnMgx4ah5cLKBavQf-qX5CLniLFVllrEbCJ9uToVgMmUqK6pz8vPy3kWX7PUbmuhmpAE_ILT0iB4DdO4bohv8Ld31Yxhm5490xGCH0IM3SAdL7xFH-orOL6bBHn2kNULwp-piURjHzpl1ZaLO02kJOkxgmjDSGKBd9AU5s9BNePl_N-Tw-HDYPif1_mm3vasTyIsqEbwUTNtCQg6gFePQSqm5ApPbttSiAiYV6KI0vORVpXImpeVSieV3yyyTG3L9N7tiaMbgeghfzQlHs-KQv-ZYXWo</recordid><startdate>20220309</startdate><enddate>20220309</enddate><creator>Cornalba, Federico</creator><creator>Disselkamp, Constantin</creator><creator>Scassola, Davide</creator><creator>Helf, Christopher</creator><scope>AKY</scope><scope>GOX</scope></search><sort><creationdate>20220309</creationdate><title>Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading</title><author>Cornalba, Federico ; Disselkamp, Constantin ; Scassola, Davide ; Helf, Christopher</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-LOGICAL-a679-21820bf73a6aab401ad33b14ac6fd8b29a034ab78c1819946033f1342485f0f03</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2022</creationdate><topic>Computer Science - Learning</topic><topic>Quantitative Finance - Computational Finance</topic><topic>Quantitative Finance - Trading and Microstructure</topic><toplevel>online_resources</toplevel><creatorcontrib>Cornalba, Federico</creatorcontrib><creatorcontrib>Disselkamp, Constantin</creatorcontrib><creatorcontrib>Scassola, Davide</creatorcontrib><creatorcontrib>Helf, Christopher</creatorcontrib><collection>arXiv Computer Science</collection><collection>arXiv.org</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext_linktorsrc</fulltext></delivery><addata><au>Cornalba, Federico</au><au>Disselkamp, Constantin</au><au>Scassola, Davide</au><au>Helf, Christopher</au><format>journal</format><genre>article</genre><ristype>JOUR</ristype><atitle>Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading</atitle><date>2022-03-09</date><risdate>2022</risdate><abstract>We investigate the potential of Multi-Objective, Deep Reinforcement Learning for stock and cryptocurrency single-asset trading: in particular, we consider a Multi-Objective algorithm which generalizes the reward functions and discount factor (i.e., these components are not specified a priori, but incorporated in the learning process). Firstly, using several important assets (cryptocurrency pairs BTCUSD, ETHUSDT, XRPUSDT, and stock indexes AAPL, SPY, NIFTY50), we verify the reward generalization property of the proposed Multi-Objective algorithm, and provide preliminary statistical evidence showing increased predictive stability over the corresponding Single-Objective strategy. Secondly, we show that the Multi-Objective algorithm has a clear edge over the corresponding Single-Objective strategy when the reward mechanism is sparse (i.e., when non-null feedback is infrequent over time). Finally, we discuss the generalization properties with respect to the discount factor. The entirety of our code is provided in open source format.</abstract><doi>10.48550/arxiv.2203.04579</doi><oa>free_for_read</oa></addata></record>
fulltext fulltext_linktorsrc
identifier DOI: 10.48550/arxiv.2203.04579
ispartof
issn
language eng
recordid cdi_arxiv_primary_2203_04579
source arXiv.org
subjects Computer Science - Learning
Quantitative Finance - Computational Finance
Quantitative Finance - Trading and Microstructure
title Multi-Objective reward generalization: Improving performance of Deep Reinforcement Learning for applications in single-asset trading
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-28T23%3A49%3A05IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-arxiv_GOX&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.genre=article&rft.atitle=Multi-Objective%20reward%20generalization:%20Improving%20performance%20of%20Deep%20Reinforcement%20Learning%20for%20applications%20in%20single-asset%20trading&rft.au=Cornalba,%20Federico&rft.date=2022-03-09&rft_id=info:doi/10.48550/arxiv.2203.04579&rft_dat=%3Carxiv_GOX%3E2203_04579%3C/arxiv_GOX%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_id=info:pmid/&rfr_iscdi=true