An application of deep reinforcement learning to algorithmic trading

•Reinforcement learning (RL) formalization of the algorithmic trading problem.•Novel trading strategy based on deep reinforcement learning (DRL), denominated TDQN.•Rigorous performance assessment methodology for algorithmic trading.•TDQN algorithm delivers promising results surpassing benchmark stra...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	Expert systems with applications Ročník 173; s. 114632
Hlavní autoři:	Théate, Thibaut, Ernst, Damien
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York Elsevier Ltd 01.07.2021 Elsevier BV Elsevier
Témata:	Algorithmic trading Algorithms Artificial intelligence Business & economic sciences Computer science Deep learning Deep reinforcement learning Engineering, computing & technology Finance Ingénierie, informatique & technologie Machine learning Performance assessment Program trading Sciences informatiques Sciences économiques & de gestion Scientific papers Securities markets Stock exchanges Trading policy Algorithmic trading Trading policy Deep reinforcement learning Artificial intelligence
ISSN:	0957-4174, 1873-6793, 1873-6793
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Abstract	•Reinforcement learning (RL) formalization of the algorithmic trading problem.•Novel trading strategy based on deep reinforcement learning (DRL), denominated TDQN.•Rigorous performance assessment methodology for algorithmic trading.•TDQN algorithm delivers promising results surpassing benchmark strategies. This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in the stock market. It proposes a novel DRL trading policy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new DRL approach is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN algorithm.
AbstractList	•Reinforcement learning (RL) formalization of the algorithmic trading problem.•Novel trading strategy based on deep reinforcement learning (DRL), denominated TDQN.•Rigorous performance assessment methodology for algorithmic trading.•TDQN algorithm delivers promising results surpassing benchmark strategies. This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in the stock market. It proposes a novel DRL trading policy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new DRL approach is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN algorithm. This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of determining the optimal trading position at any point in time during a trading activity in the stock market. It proposes a novel DRL trading policy so as to maximise the resulting Sharpe ratio performance indicator on a broad range of stock markets. Denominated the Trading Deep Q-Network algorithm (TDQN), this new DRL approach is inspired from the popular DQN algorithm and significantly adapted to the specific algorithmic trading problem at hand. The training of the resulting reinforcement learning (RL) agent is entirely based on the generation of artificial trajectories from a limited set of stock market historical data. In order to objectively assess the performance of trading strategies, the research paper also proposes a novel, more rigorous performance assessment methodology. Following this new performance assessment approach, promising results are reported for the TDQN algorithm.
ArticleNumber	114632
Author	Ernst, Damien Théate, Thibaut
Author_xml	– sequence: 1 givenname: Thibaut surname: Théate fullname: Théate, Thibaut email: thibaut.theate@uliege.be – sequence: 2 givenname: Damien surname: Ernst fullname: Ernst, Damien email: dernst@uliege.be
BookMark	eNp9kE1P3DAQhq2KSl2gf6CnSJyz2BPbSSQuiJYPCYkLnEdee7x4lbVTx0vVf98s6YkDpxmN3mc085yyk5giMfZD8LXgQl_u1jT9MWvgINZCSN3AF7YSXdvUuu2bE7bivWprKVr5jZ1O045z0XLertjP61iZcRyCNSWkWCVfOaKxyhSiT9nSnmKpBjI5hritSqrMsE05lNd9sFXJxs3jc_bVm2Gi7__rGXu5_fV8c18_Pt093Fw_1lYKXWqSRoHiUvbOawLlvQGtrJQbB411ThhqPJDwjVLGK6kdtWrTASkDZHtozliz7B0CbQlT3gR8A0wmLP1h2KKxuCEE0B2C1FK1M3WxUGNOvw80FdylQ47zoQgKhBDQ9XpOdUvK5jRNmTzaUN6dzE-GAQXHo2nc4dE0Hk3jYnpG4QM65rA3-e_n0NUC0SzsLVDGyQaKllzIZAu6FD7D_wFXiJlo
CitedBy_id	crossref_primary_10_1016_j_asoc_2021_107788 crossref_primary_10_3390_electronics13234659 crossref_primary_10_1051_e3sconf_202337605002 crossref_primary_10_3390_en13236435 crossref_primary_10_3390_en15103645 crossref_primary_10_1007_s13177_025_00521_9 crossref_primary_10_1016_j_procs_2021_09_147 crossref_primary_10_3390_data6110119 crossref_primary_10_1007_s41060_024_00692_w crossref_primary_10_1016_j_asoc_2025_113029 crossref_primary_10_1109_ACCESS_2025_3546099 crossref_primary_10_1007_s00500_023_08973_5 crossref_primary_10_1007_s42979_024_03555_0 crossref_primary_10_1145_3733714 crossref_primary_10_1016_j_eswa_2023_120346 crossref_primary_10_1016_j_neunet_2025_107905 crossref_primary_10_1016_j_eswa_2023_121711 crossref_primary_10_1109_ACCESS_2025_3558887 crossref_primary_10_1016_j_cirp_2024_06_003 crossref_primary_10_1109_ACCESS_2023_3289844 crossref_primary_10_1109_ACCESS_2024_3418510 crossref_primary_10_1177_18724981251315846 crossref_primary_10_3390_app122412526 crossref_primary_10_1016_j_jenvman_2024_123308 crossref_primary_10_1103_PhysRevApplied_21_067001 crossref_primary_10_1016_j_eswa_2025_127864 crossref_primary_10_1155_2021_7877590 crossref_primary_10_1007_s10489_022_03280_2 crossref_primary_10_1016_j_knosys_2024_112767 crossref_primary_10_3390_ai6080183 crossref_primary_10_1002_jsc_2525 crossref_primary_10_1016_j_asoc_2025_113881 crossref_primary_10_1111_exsy_13667 crossref_primary_10_3390_app13031956 crossref_primary_10_3390_electronics11091506 crossref_primary_10_1007_s00521_020_05377_6 crossref_primary_10_1038_s41598_024_51408_w crossref_primary_10_1016_j_eswa_2022_118614 crossref_primary_10_3390_a16010023 crossref_primary_10_1007_s13762_023_04763_6 crossref_primary_10_3390_math12244020 crossref_primary_10_3390_info15120755 crossref_primary_10_1007_s10462_023_10512_5 crossref_primary_10_1016_j_asoc_2023_110802 crossref_primary_10_1016_j_eswa_2023_121245 crossref_primary_10_1016_j_aej_2025_02_013 crossref_primary_10_1016_j_eswa_2023_122581 crossref_primary_10_1109_ACCESS_2024_3515039 crossref_primary_10_3390_risks13080148 crossref_primary_10_1109_ACCESS_2022_3226629 crossref_primary_10_1146_annurev_statistics_112723_034423 crossref_primary_10_1016_j_asoc_2025_113252 crossref_primary_10_1109_ACCESS_2023_3253503 crossref_primary_10_3390_math11173626 crossref_primary_10_3390_a17080343 crossref_primary_10_3390_ijfs11010010 crossref_primary_10_1016_j_eswa_2022_117311 crossref_primary_10_1007_s10462_024_11066_w crossref_primary_10_3390_a16070325 crossref_primary_10_1016_j_eswa_2023_121897 crossref_primary_10_1016_j_knosys_2023_111290 crossref_primary_10_1016_j_asoc_2023_111108 crossref_primary_10_1016_j_jksuci_2024_102015 crossref_primary_10_5772_acrt_20230095 crossref_primary_10_1007_s10489_022_03321_w crossref_primary_10_1016_j_asoc_2025_113927 crossref_primary_10_1177_20438869231189519 crossref_primary_10_3390_math13152485 crossref_primary_10_1016_j_engappai_2023_107004 crossref_primary_10_3390_app112311208 crossref_primary_10_3390_systems10050146 crossref_primary_10_1109_ACCESS_2023_3259424 crossref_primary_10_1016_j_jisa_2024_103856 crossref_primary_10_1016_j_apenergy_2023_121321 crossref_primary_10_1016_j_asoc_2025_113617 crossref_primary_10_3390_risks13030040 crossref_primary_10_3390_s23167216 crossref_primary_10_1016_j_eswa_2023_121502 crossref_primary_10_1631_FITEE_2200039 crossref_primary_10_1016_j_eswa_2023_122994 crossref_primary_10_1155_2022_6820073 crossref_primary_10_21511_imfi_21_1__2024_21 crossref_primary_10_1016_j_eswa_2024_124350 crossref_primary_10_1016_j_eswa_2024_125043 crossref_primary_10_1016_j_eswa_2025_128297
Cites_doi	10.1016/j.asoc.2018.09.017 10.1090/noti1105 10.1016/j.eswa.2018.08.003 10.1109/72.935097 10.1016/j.eswa.2005.10.012 10.1111/j.1540-6261.2010.01624.x 10.1016/j.eswa.2020.113573 10.1109/MSP.2017.2743240 10.1007/s10994-021-06020-8 10.1038/nature14236 10.1109/TNNLS.2016.2522401 10.1371/journal.pmed.0020124 10.1109/TKDE.2013.133 10.1109/MC.2011.31 10.1371/journal.pone.0180944 10.1016/j.eswa.2019.112872 10.1016/j.eswa.2018.09.036 10.1016/j.eswa.2019.04.013 10.3905/jpm.2011.38.1.110 10.1609/aaai.v32i1.11796 10.1145/2500117 10.1038/nature14539 10.1016/j.jocs.2010.12.007 10.1038/nature16961
ContentType	Journal Article
Contributor	Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège
Contributor_xml	– sequence: 1 fullname: Montefiore Institute - Montefiore Institute of Electrical Engineering and Computer Science - ULiège
Copyright	2021 Elsevier Ltd Copyright Elsevier BV Jul 1, 2021
Copyright_xml	– notice: 2021 Elsevier Ltd – notice: Copyright Elsevier BV Jul 1, 2021
DBID	AAYXX CITATION 7SC 8FD JQ2 L7M L~C L~D JLOSS Q33
DOI	10.1016/j.eswa.2021.114632
DatabaseName	CrossRef Computer and Information Systems Abstracts Technology Research Database ProQuest Computer Science Collection Advanced Technologies Database with Aerospace Computer and Information Systems Abstracts Academic Computer and Information Systems Abstracts Professional Université de Liège - Open Repository and Bibliography (ORBI) (Open Access titles only) Université de Liège - Open Repository and Bibliography (ORBI)
DatabaseTitle	CrossRef Computer and Information Systems Abstracts Technology Research Database Computer and Information Systems Abstracts – Academic Advanced Technologies Database with Aerospace ProQuest Computer Science Collection Computer and Information Systems Abstracts Professional
DatabaseTitleList	Computer and Information Systems Abstracts
DeliveryMethod	fulltext_linktorsrc
Discipline	Computer Science
EISSN	1873-6793
ExternalDocumentID	oai_orbi_ulg_ac_be_2268_246457 10_1016_j_eswa_2021_114632 S0957417421000737
GroupedDBID	--K --M .DC .~1 0R~ 13V 1B1 1RT 1~. 1~5 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JN 9JO AAAKF AABNK AACTN AAEDT AAEDW AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AARIN AAXUO AAYFN ABBOA ABFNM ABMAC ABMVD ABUCO ABYKQ ACDAQ ACGFS ACHRH ACNTT ACRLP ACZNC ADBBV ADEZE ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGJBL AGUBO AGUMN AGYEJ AHHHB AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJOXV ALEQD ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD APLSM AXJTR BJAXD BKOJK BLXMC BNSAS CS3 DU5 EBS EFJIC EFLBG EO8 EO9 EP2 EP3 F5P FDB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HAMUX IHE J1W JJJVA KOM LG9 LY1 LY7 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. PQQKQ Q38 ROL RPZ SDF SDG SDP SDS SES SPC SPCBC SSB SSD SSL SST SSV SSZ T5K TN5 ~G- 29G 9DU AAAKG AAQXK AATTM AAXKI AAYWO AAYXX ABJNI ABKBG ABUFD ABWVN ABXDB ACLOT ACNNM ACRPL ACVFH ADCNI ADJOM ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EJD FEDTE FGOYB G-2 HLZ HVGLF HZ~ R2- SBC SET SEW WUQ XPP ZMT ~HD 7SC 8FD JQ2 L7M L~C L~D JLOSS Q33
ID	FETCH-LOGICAL-c416t-e4a5250449df6e25ffa265c44bd23cdd1ae3f2e1f355af546de75b82e5a2ec923
ISICitedReferencesCount	116
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000636782600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0957-4174 1873-6793
IngestDate	Sat Nov 29 01:28:25 EST 2025 Sun Nov 09 06:43:39 EST 2025 Tue Nov 18 21:07:07 EST 2025 Sat Nov 29 07:09:02 EST 2025 Fri Feb 23 02:43:49 EST 2024
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Keywords	Algorithmic trading Trading policy Deep reinforcement learning Artificial intelligence
Language	English
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c416t-e4a5250449df6e25ffa265c44bd23cdd1ae3f2e1f355af546de75b82e5a2ec923
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14 scopus-id:2-s2.0-85101170161
ORCID	0000-0002-3035-8260
OpenAccessLink	https://orbi.uliege.be/handle/2268/246457
PQID	2521112896
PQPubID	2045477
ParticipantIDs	liege_orbi_v2_oai_orbi_ulg_ac_be_2268_246457 proquest_journals_2521112896 crossref_citationtrail_10_1016_j_eswa_2021_114632 crossref_primary_10_1016_j_eswa_2021_114632 elsevier_sciencedirect_doi_10_1016_j_eswa_2021_114632
PublicationCentury	2000
PublicationDate	2021-07-01
PublicationDateYYYYMMDD	2021-07-01
PublicationDate_xml	– month: 07 year: 2021 text: 2021-07-01 day: 01
PublicationDecade	2020
PublicationPlace	New York
PublicationPlace_xml	– name: New York
PublicationTitle	Expert systems with applications
PublicationYear	2021
Publisher	Elsevier Ltd Elsevier BV Elsevier
Publisher_xml	– name: Elsevier Ltd – name: Elsevier BV – name: Elsevier
References	Arévalo, Niño, Hernández, Sandoval (b0010) 2016 Li, Y. (2017). Deep reinforcement learning: An overview. CoRR, abs/1701.07274. Leinweber, Sisk (b0135) 2011; 38 Bailey, Borwein, de Prado, Zhu (b0020) 2014 Lei, K., Zhang, B., Li, Y., Yang, M., & Shen, Y. (2020). Time-driven Feature-aware jointly deep reinforcement learning for financial signal representation and algorithmic trading. Sutton, Barto (b0205) 2018 Boukas, I., Ernst, D., Théate, T., Bolland, A., Huynen, A., Buchwald, M., Wynants, C., & Cornélusse, B. (2020). A deep reinforcement learning framework for continuous intraday market bidding. ArXiv, abs/2004.05940. Hessel, M., Modayil, J., van Hasselt, H. P., Schaul, T., Ostrovski, G., Dabney, W., Horgan, D., Piot, B., Azar, M. G., & Silver, D. (2017). Rainbow: Combining improvements in deep reinforcement learning. CoRR, abs/1710.02298. Goodfellow, Bengio, Courville (b0080) 2016 Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. CoRR, abs/1412.6980. Watkins, Dayan (b0230) 1992; 8 Busoniu, Babuska, De Schutter, Ernst (b0045) 2010 Park, Sim, Choi (b0180) 2020; 158 Dempster, Leemans (b0065) 2006; 30 van Hasselt, H. P., Guez, A., & Silver, D. (2015). Deep reinforcement learning with double Q-Learning. CoRR, abs/1509.06461. Paiva, Cardoso, Hanaoka, Duarte (b0175) 2019; 115 . Chan (b0055) 2009 Bao, Yue, Rao (b0025) 2017; 12 Bellemare, M. G., Dabney, W., & Munos, R. (2017). A distributional perspective on reinforcement learning. CoRR, abs/1707.06887. Chan (b0060) 2013 Wang, Z., de Freitas, N., & Lanctot, M. (2015). Dueling network architectures for deep reinforcement learning. CoRR, abs/1511.06581. Hausknecht, M. J., & Stone, P. (2015). Deep recurrent Q-Learning for partially observable MDPs. CoRR, abs/1507.06527. Nuti, Mirghaemi, Treleaven, Yingsaeree (b0170) 2011; 44 Zhang, C., Vinyals, O., Munos, R., & Bengio, S. (2018). A study on overfitting in deep reinforcement learning. CoRR, abs/1804.06893. Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017). Proximal policy optimization algorithms. CoRR, abs/1707.06347. Shao, K., Tang, Z., Zhu, Y., Li, N., & Zhao, D. (2019). A survey of deep reinforcement learning in video games. ArXiv, abs/1912.10944. Hendershott, Jones, Menkveld (b0095) 2011; 66 Mnih, Kavukcuoglu, Silver, Rusu, Veness, Bellemare, Graves, Riedmiller, Fidjeland, Ostrovski, Petersen, Beattie, Sadik, Antonoglou, King, Kumaran, Wierstra, Legg, Hassabis (b0150) 2015; 518 Fortunato, M., Azar, M. G., Piot, B., Menick, J., Hessel, M., Osband, I., Graves, A., Mnih, V., Munos, R., Hassabis, D., Pietquin, O., Blundell, C., & Legg, S. (2018). Noisy networks for exploration. CoRR, abs/1706.10295. Ioffe, S., & Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. CoRR, abs/1502.03167. Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. A. (2013). Playing Atari with deep reinforcement learning. CoRR, abs/1312.5602. Almahdi, Yang (b0005) 2019; 130 Arulkumaran, K., Deisenroth, M. P., Brundage, M., & Bharath, A. A. (2017). A brief survey of deep reinforcement learning. CoRR, abs/1708.05866. Carapuço, Neves, Horta (b0050) 2018; 73 Treleaven, Galas, Lalchand (b0215) 2013; 56 Nuij, Milea, Hogenboom, Frasincar, Kaymak (b0165) 2014; 26 Goodfellow, Bengio, Courville (b0085) 2015; 521 Silver, Huang, Maddison, Guez, Sifre, van den Driessche, Schrittwieser, Antonoglou, Panneershelvam, Lanctot, Dieleman, Grewe, Nham, Kalchbrenner, Sutskever, Lillicrap, Leach, Kavukcuoglu, Graepel, Hassabis (b0200) 2016; 529 Deng, Bao, Kong, Ren, Dai (b0070) 2017; 28 Szepesvari (b0210) 2010 Bollen, J., Mao, H., & jun Zeng, X. (2011). Twitter mood predicts the stock market. LeCun, Bengio, Hinton (b0125) 2015; 521 1–8. Schaul, T., Quan, J., Antonoglou, I., & Silver, D. (2016). Prioritized experience replay. CoRR, abs/1511.05952. Jeong, Kim (b0115) 2019; 117 Moody, Saffell (b0155) 2001; 12 Ioannidis (b0105) 2005; 2 Narang (b0160) 2009 Leinweber (10.1016/j.eswa.2021.114632_b0135) 2011; 38 Chan (10.1016/j.eswa.2021.114632_b0060) 2013 Paiva (10.1016/j.eswa.2021.114632_b0175) 2019; 115 10.1016/j.eswa.2021.114632_b0040 Bailey (10.1016/j.eswa.2021.114632_b0020) 2014 10.1016/j.eswa.2021.114632_b0220 10.1016/j.eswa.2021.114632_b0120 10.1016/j.eswa.2021.114632_b0185 10.1016/j.eswa.2021.114632_b0140 Szepesvari (10.1016/j.eswa.2021.114632_b0210) 2010 10.1016/j.eswa.2021.114632_b0015 10.1016/j.eswa.2021.114632_b0235 10.1016/j.eswa.2021.114632_b0035 LeCun (10.1016/j.eswa.2021.114632_b0125) 2015; 521 Jeong (10.1016/j.eswa.2021.114632_b0115) 2019; 117 Busoniu (10.1016/j.eswa.2021.114632_b0045) 2010 Goodfellow (10.1016/j.eswa.2021.114632_b0085) 2015; 521 Watkins (10.1016/j.eswa.2021.114632_b0230) 1992; 8 Mnih (10.1016/j.eswa.2021.114632_b0150) 2015; 518 Nuij (10.1016/j.eswa.2021.114632_b0165) 2014; 26 10.1016/j.eswa.2021.114632_b0190 Arévalo (10.1016/j.eswa.2021.114632_b0010) 2016 Carapuço (10.1016/j.eswa.2021.114632_b0050) 2018; 73 10.1016/j.eswa.2021.114632_b0090 Goodfellow (10.1016/j.eswa.2021.114632_b0080) 2016 Ioannidis (10.1016/j.eswa.2021.114632_b0105) 2005; 2 10.1016/j.eswa.2021.114632_b0110 Almahdi (10.1016/j.eswa.2021.114632_b0005) 2019; 130 Sutton (10.1016/j.eswa.2021.114632_b0205) 2018 10.1016/j.eswa.2021.114632_b0075 10.1016/j.eswa.2021.114632_b0130 10.1016/j.eswa.2021.114632_b0030 10.1016/j.eswa.2021.114632_b0195 Nuti (10.1016/j.eswa.2021.114632_b0170) 2011; 44 Bao (10.1016/j.eswa.2021.114632_b0025) 2017; 12 Narang (10.1016/j.eswa.2021.114632_b0160) 2009 10.1016/j.eswa.2021.114632_b0145 10.1016/j.eswa.2021.114632_b0100 Silver (10.1016/j.eswa.2021.114632_b0200) 2016; 529 Hendershott (10.1016/j.eswa.2021.114632_b0095) 2011; 66 10.1016/j.eswa.2021.114632_b0225 Treleaven (10.1016/j.eswa.2021.114632_b0215) 2013; 56 Moody (10.1016/j.eswa.2021.114632_b0155) 2001; 12 Park (10.1016/j.eswa.2021.114632_b0180) 2020; 158 Chan (10.1016/j.eswa.2021.114632_b0055) 2009 Deng (10.1016/j.eswa.2021.114632_b0070) 2017; 28 Dempster (10.1016/j.eswa.2021.114632_b0065) 2006; 30
References_xml	– volume: 28 start-page: 653 year: 2017 end-page: 664 ident: b0070 article-title: Deep direct reinforcement learning for financial signal representation and trading publication-title: IEEE Transactions on Neural Networks and Learning Systems – reference: Schaul, T., Quan, J., Antonoglou, I., & Silver, D. (2016). Prioritized experience replay. CoRR, abs/1511.05952. – volume: 2 start-page: 124 year: 2005 ident: b0105 article-title: Why most published research findings are false publication-title: PLoS Med – volume: 12 year: 2017 ident: b0025 article-title: A deep learning framework for financial time series using stacked autoencoders and long-short term memory publication-title: PloS one – volume: 529 start-page: 484 year: 2016 end-page: 489 ident: b0200 article-title: Mastering the game of go with deep neural networks and tree search publication-title: Nature – reference: Bollen, J., Mao, H., & jun Zeng, X. (2011). Twitter mood predicts the stock market. – volume: 521 year: 2015 ident: b0125 article-title: Deep learning publication-title: Nature – start-page: 458 year: 2014 end-page: 471 ident: b0020 article-title: Pseudo-mathematics and financial charlatanism: the effects of backtest overfitting on out-of-sample performance publication-title: Notice of the American Mathematical Society – year: 2010 ident: b0045 article-title: Reinforcement learning and dynamic programming using function approximators – volume: 158 year: 2020 ident: b0180 article-title: An intelligent financial portfolio trading strategy using Deep Q-Learning publication-title: Expert Systems With Applications – reference: Zhang, C., Vinyals, O., Munos, R., & Bengio, S. (2018). A study on overfitting in deep reinforcement learning. CoRR, abs/1804.06893. – reference: Lei, K., Zhang, B., Li, Y., Yang, M., & Shen, Y. (2020). Time-driven Feature-aware jointly deep reinforcement learning for financial signal representation and algorithmic trading. – reference: Hessel, M., Modayil, J., van Hasselt, H. P., Schaul, T., Ostrovski, G., Dabney, W., Horgan, D., Piot, B., Azar, M. G., & Silver, D. (2017). Rainbow: Combining improvements in deep reinforcement learning. CoRR, abs/1710.02298. – volume: 8 start-page: 279 year: 1992 end-page: 292 ident: b0230 article-title: Technical note: Q-Learning publication-title: Machine Learning – reference: Bellemare, M. G., Dabney, W., & Munos, R. (2017). A distributional perspective on reinforcement learning. CoRR, abs/1707.06887. – volume: 117 start-page: 125 year: 2019 end-page: 138 ident: b0115 article-title: Improving financial trading decisions using Deep Q-Learning: Predicting the number of shares, action strategies, and transfer learning publication-title: Expert Systems With Applications – reference: Hausknecht, M. J., & Stone, P. (2015). Deep recurrent Q-Learning for partially observable MDPs. CoRR, abs/1507.06527. – reference: Li, Y. (2017). Deep reinforcement learning: An overview. CoRR, abs/1701.07274. – year: 2010 ident: b0210 article-title: Algorithms for reinforcement learning – reference: Arulkumaran, K., Deisenroth, M. P., Brundage, M., & Bharath, A. A. (2017). A brief survey of deep reinforcement learning. CoRR, abs/1708.05866. – volume: 518 start-page: 529 year: 2015 end-page: 533 ident: b0150 article-title: Human-level control through deep reinforcement learning publication-title: Nature – year: 2016 ident: b0010 article-title: High-frequency trading strategy based on deep neural networks publication-title: ICIC – volume: 521 start-page: 436 year: 2015 end-page: 444 ident: b0085 article-title: Deep learning publication-title: Nature – reference: Ioffe, S., & Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. CoRR, abs/1502.03167. – volume: 130 start-page: 145 year: 2019 end-page: 156 ident: b0005 article-title: A constrained portfolio trading system using particle swarm algorithm and recurrent reinforcement learning publication-title: Expert Systems With Applications – reference: Shao, K., Tang, Z., Zhu, Y., Li, N., & Zhao, D. (2019). A survey of deep reinforcement learning in video games. ArXiv, abs/1912.10944. – volume: 26 start-page: 823 year: 2014 end-page: 835 ident: b0165 article-title: An automated framework for incorporating news into stock trading strategies publication-title: IEEE Transactions on Knowledge and Data Engineering – volume: 56 start-page: 76 year: 2013 end-page: 85 ident: b0215 article-title: Algorithmic trading review publication-title: Communications of the ACM – volume: 66 start-page: 1 year: 2011 end-page: 33 ident: b0095 article-title: Does algorithmic trading improve liquidity? publication-title: Journal of Finance – volume: 12 start-page: 875 year: 2001 end-page: 889 ident: b0155 article-title: Learning to trade via direct reinforcement publication-title: IEEE Transactions on Neural Networks – reference: Schulman, J., Wolski, F., Dhariwal, P., Radford, A., & Klimov, O. (2017). Proximal policy optimization algorithms. CoRR, abs/1707.06347. – volume: 73 start-page: 783 year: 2018 end-page: 794 ident: b0050 article-title: Reinforcement learning applied to Forex trading publication-title: Applied Soft Computing – volume: 115 start-page: 635 year: 2019 end-page: 655 ident: b0175 article-title: Decision-making for financial trading: A fusion approach of machine learning and portfolio selection publication-title: Expert Systems With Applications – reference: Wang, Z., de Freitas, N., & Lanctot, M. (2015). Dueling network architectures for deep reinforcement learning. CoRR, abs/1511.06581. – reference: Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. CoRR, abs/1412.6980. – reference: Mnih, V., Kavukcuoglu, K., Silver, D., Graves, A., Antonoglou, I., Wierstra, D., & Riedmiller, M. A. (2013). Playing Atari with deep reinforcement learning. CoRR, abs/1312.5602. – volume: 30 start-page: 543 year: 2006 end-page: 552 ident: b0065 article-title: An automated FX trading system using adaptive reinforcement learning publication-title: Expert Systems With Applications – reference: Fortunato, M., Azar, M. G., Piot, B., Menick, J., Hessel, M., Osband, I., Graves, A., Mnih, V., Munos, R., Hassabis, D., Pietquin, O., Blundell, C., & Legg, S. (2018). Noisy networks for exploration. CoRR, abs/1706.10295. – volume: 38 start-page: 110 year: 2011 end-page: 124 ident: b0135 article-title: Event-driven trading and the “New News” publication-title: The Journal of Portfolio Management – year: 2013 ident: b0060 article-title: Algorithmic trading: winning strategies and their rationale – year: 2018 ident: b0205 article-title: Reinforcement learning: An introduction – year: 2009 ident: b0055 article-title: Quantitative trading: how to build your own algorithmic trading business – reference: . – volume: 44 start-page: 61 year: 2011 end-page: 69 ident: b0170 article-title: Algorithmic trading publication-title: Computer – reference: , 1–8. – reference: van Hasselt, H. P., Guez, A., & Silver, D. (2015). Deep reinforcement learning with double Q-Learning. CoRR, abs/1509.06461. – year: 2009 ident: b0160 article-title: Inside the black box – year: 2016 ident: b0080 article-title: Deep learning – reference: Boukas, I., Ernst, D., Théate, T., Bolland, A., Huynen, A., Buchwald, M., Wynants, C., & Cornélusse, B. (2020). A deep reinforcement learning framework for continuous intraday market bidding. ArXiv, abs/2004.05940. – volume: 73 start-page: 783 year: 2018 ident: 10.1016/j.eswa.2021.114632_b0050 article-title: Reinforcement learning applied to Forex trading publication-title: Applied Soft Computing doi: 10.1016/j.asoc.2018.09.017 – year: 2016 ident: 10.1016/j.eswa.2021.114632_b0010 article-title: High-frequency trading strategy based on deep neural networks – start-page: 458 year: 2014 ident: 10.1016/j.eswa.2021.114632_b0020 article-title: Pseudo-mathematics and financial charlatanism: the effects of backtest overfitting on out-of-sample performance publication-title: Notice of the American Mathematical Society doi: 10.1090/noti1105 – volume: 115 start-page: 635 year: 2019 ident: 10.1016/j.eswa.2021.114632_b0175 article-title: Decision-making for financial trading: A fusion approach of machine learning and portfolio selection publication-title: Expert Systems With Applications doi: 10.1016/j.eswa.2018.08.003 – year: 2010 ident: 10.1016/j.eswa.2021.114632_b0210 – volume: 12 start-page: 875 issue: 4 year: 2001 ident: 10.1016/j.eswa.2021.114632_b0155 article-title: Learning to trade via direct reinforcement publication-title: IEEE Transactions on Neural Networks doi: 10.1109/72.935097 – volume: 30 start-page: 543 year: 2006 ident: 10.1016/j.eswa.2021.114632_b0065 article-title: An automated FX trading system using adaptive reinforcement learning publication-title: Expert Systems With Applications doi: 10.1016/j.eswa.2005.10.012 – volume: 66 start-page: 1 year: 2011 ident: 10.1016/j.eswa.2021.114632_b0095 article-title: Does algorithmic trading improve liquidity? publication-title: Journal of Finance doi: 10.1111/j.1540-6261.2010.01624.x – volume: 158 year: 2020 ident: 10.1016/j.eswa.2021.114632_b0180 article-title: An intelligent financial portfolio trading strategy using Deep Q-Learning publication-title: Expert Systems With Applications doi: 10.1016/j.eswa.2020.113573 – ident: 10.1016/j.eswa.2021.114632_b0185 – ident: 10.1016/j.eswa.2021.114632_b0015 doi: 10.1109/MSP.2017.2743240 – ident: 10.1016/j.eswa.2021.114632_b0040 doi: 10.1007/s10994-021-06020-8 – ident: 10.1016/j.eswa.2021.114632_b0140 – volume: 518 start-page: 529 year: 2015 ident: 10.1016/j.eswa.2021.114632_b0150 article-title: Human-level control through deep reinforcement learning publication-title: Nature doi: 10.1038/nature14236 – volume: 28 start-page: 653 year: 2017 ident: 10.1016/j.eswa.2021.114632_b0070 article-title: Deep direct reinforcement learning for financial signal representation and trading publication-title: IEEE Transactions on Neural Networks and Learning Systems doi: 10.1109/TNNLS.2016.2522401 – ident: 10.1016/j.eswa.2021.114632_b0110 – ident: 10.1016/j.eswa.2021.114632_b0225 – year: 2018 ident: 10.1016/j.eswa.2021.114632_b0205 – volume: 2 start-page: 124 year: 2005 ident: 10.1016/j.eswa.2021.114632_b0105 article-title: Why most published research findings are false publication-title: PLoS Med doi: 10.1371/journal.pmed.0020124 – ident: 10.1016/j.eswa.2021.114632_b0190 – volume: 26 start-page: 823 year: 2014 ident: 10.1016/j.eswa.2021.114632_b0165 article-title: An automated framework for incorporating news into stock trading strategies publication-title: IEEE Transactions on Knowledge and Data Engineering doi: 10.1109/TKDE.2013.133 – volume: 44 start-page: 61 year: 2011 ident: 10.1016/j.eswa.2021.114632_b0170 article-title: Algorithmic trading publication-title: Computer doi: 10.1109/MC.2011.31 – volume: 12 year: 2017 ident: 10.1016/j.eswa.2021.114632_b0025 article-title: A deep learning framework for financial time series using stacked autoencoders and long-short term memory publication-title: PloS one doi: 10.1371/journal.pone.0180944 – year: 2013 ident: 10.1016/j.eswa.2021.114632_b0060 – ident: 10.1016/j.eswa.2021.114632_b0130 doi: 10.1016/j.eswa.2019.112872 – ident: 10.1016/j.eswa.2021.114632_b0090 – volume: 117 start-page: 125 year: 2019 ident: 10.1016/j.eswa.2021.114632_b0115 article-title: Improving financial trading decisions using Deep Q-Learning: Predicting the number of shares, action strategies, and transfer learning publication-title: Expert Systems With Applications doi: 10.1016/j.eswa.2018.09.036 – volume: 130 start-page: 145 year: 2019 ident: 10.1016/j.eswa.2021.114632_b0005 article-title: A constrained portfolio trading system using particle swarm algorithm and recurrent reinforcement learning publication-title: Expert Systems With Applications doi: 10.1016/j.eswa.2019.04.013 – volume: 38 start-page: 110 year: 2011 ident: 10.1016/j.eswa.2021.114632_b0135 article-title: Event-driven trading and the “New News” publication-title: The Journal of Portfolio Management doi: 10.3905/jpm.2011.38.1.110 – ident: 10.1016/j.eswa.2021.114632_b0235 – ident: 10.1016/j.eswa.2021.114632_b0100 doi: 10.1609/aaai.v32i1.11796 – ident: 10.1016/j.eswa.2021.114632_b0120 – year: 2010 ident: 10.1016/j.eswa.2021.114632_b0045 – year: 2016 ident: 10.1016/j.eswa.2021.114632_b0080 – ident: 10.1016/j.eswa.2021.114632_b0145 – ident: 10.1016/j.eswa.2021.114632_b0075 – volume: 56 start-page: 76 year: 2013 ident: 10.1016/j.eswa.2021.114632_b0215 article-title: Algorithmic trading review publication-title: Communications of the ACM doi: 10.1145/2500117 – volume: 521 start-page: 436 year: 2015 ident: 10.1016/j.eswa.2021.114632_b0085 article-title: Deep learning publication-title: Nature doi: 10.1038/nature14539 – ident: 10.1016/j.eswa.2021.114632_b0035 doi: 10.1016/j.jocs.2010.12.007 – year: 2009 ident: 10.1016/j.eswa.2021.114632_b0055 – volume: 529 start-page: 484 year: 2016 ident: 10.1016/j.eswa.2021.114632_b0200 article-title: Mastering the game of go with deep neural networks and tree search publication-title: Nature doi: 10.1038/nature16961 – ident: 10.1016/j.eswa.2021.114632_b0030 – ident: 10.1016/j.eswa.2021.114632_b0220 – volume: 521 year: 2015 ident: 10.1016/j.eswa.2021.114632_b0125 article-title: Deep learning publication-title: Nature doi: 10.1038/nature14539 – volume: 8 start-page: 279 year: 1992 ident: 10.1016/j.eswa.2021.114632_b0230 article-title: Technical note: Q-Learning publication-title: Machine Learning – ident: 10.1016/j.eswa.2021.114632_b0195 – year: 2009 ident: 10.1016/j.eswa.2021.114632_b0160
SSID	ssj0017007
Score	2.6686528
Snippet	•Reinforcement learning (RL) formalization of the algorithmic trading problem.•Novel trading strategy based on deep reinforcement learning (DRL), denominated... This scientific research paper presents an innovative approach based on deep reinforcement learning (DRL) to solve the algorithmic trading problem of...
SourceID	liege proquest crossref elsevier
SourceType	Open Access Repository Aggregation Database Enrichment Source Index Database Publisher
StartPage	114632
SubjectTerms	Algorithmic trading Algorithms Artificial intelligence Business & economic sciences Computer science Deep learning Deep reinforcement learning Engineering, computing & technology Finance Ingénierie, informatique & technologie Machine learning Performance assessment Program trading Sciences informatiques Sciences économiques & de gestion Scientific papers Securities markets Stock exchanges Trading policy
Title	An application of deep reinforcement learning to algorithmic trading
URI	https://dx.doi.org/10.1016/j.eswa.2021.114632 https://www.proquest.com/docview/2521112896 https://orbi.uliege.be/handle/2268/246457
Volume	173
WOSCitedRecordID	wos000636782600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1873-6793 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0017007 issn: 0957-4174 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Nb9MwFLeg5cCFb0THQD5wg6DFcWLnGNEi4DBxKFJvlmO_dKlKUqXp2J8_O3aydhMVO3CJoiiJovd-8fv5fSL0IaSSqxTSICq4DqgswiBnlAY0Byl1CprHqhs2wc7P-WKR_vRTHLfdOAFWVfzqKt38V1Wba0bZtnT2HuoeXmoumHOjdHM0ajfHf1J8VjlqqQYyOAXYGDl2PVJV5w7s26ouLfXM1su6KdsLmyVvTNdgzFZDlh40rW_53BfD7YW9b1JLXNBduoF784syl7shqWbWVK66ZCp_l778zDsbSDgkpnoPmDfXB25EFtDQTdq5sxw7z8DqM2z_2B5PJOxaE3uH5kHv61s26aD7dd3kpditl0IqkYMwhJELYiOy7CEaExanfITG2ffZ4scQP2JnrlC-_zq77eYsChKzHPnSKZfld_vL_kZPxmubw3DHYHcsZP4MPfHbB5w5tT9HD6B6gZ72ozmwX6lfomlW4T0U4LrAFgX4AAW4RwFua7yHAuxR8Ar9-jqbf_kW-IEZgTK8ug2AShulpjTVRQIkLgpJklhRmmsSKa1DCVFBICwMyZRFTBMNLM45gVgSUIbqv0ajqq7gDcJ5ylMdnam8SGKqmeKMQWS2ugpIDqGmExT2chLKd5O3Q03Wok8bXAkrW2FlK5xsJ-jj8MzG9VI5enfci194NuhYnjDoOvrcp05XDjaXRByH0ASd9ioV_i_eCmJIrdmI8DQ5ud_b3qLHN7_NKRq1zQ7eoUfqsi23zXuP0msKWpta
linkProvider	Elsevier
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=An+Application+of+Deep+Reinforcement+Learning+to+Algorithmic+Trading&rft.jtitle=Expert+systems+with+applications&rft.au=Th%C3%A9ate%2C+Thibaut&rft.au=Ernst%2C+Damien&rft.date=2021-07-01&rft.pub=Elsevier&rft.issn=0957-4174&rft_id=info:doi/10.1016%2Fj.eswa.2021.114632&rft.externalDBID=n%2Fa&rft.externalDocID=oai_orbi_ulg_ac_be_2268_246457
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0957-4174&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0957-4174&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0957-4174&client=summon