A novel CNN-DDPG based AI-trader: Performance and roles in business operations

Luo, Suyuan; Lin, Xudong; Zheng, Zunxin

doi:10.1016/j.tre.2019.09.013

Cited by 54 publications

(15 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To deal with COVID-19, many state-of-the-arts technologies can be used. For example, using big data (Choi et al 2018a, b;Akter and Wamba 2019;Aras et al 2020), artificial intelligence (Luo et al 2019), blockchain (Choi 2019;Choi et al 2019b;Choi and Luo 2019;Cai et al 2020;Kalla et al 2020) and wireless networks (Siriwardhana et al 2020) can help to enhance traceability of patients and people suspected of having virus. Blockchain can also help to facilitate election and voting by keeping a permanent digital record and supporting secure digital operations.…”

Section: Using Information Technologies (Rd8)mentioning

confidence: 99%

Fighting against COVID-19: what operations research can help and the sense-and-respond framework

Choi

2021

Ann Oper Res

View full text Add to dashboard Cite

COVID-19 is affecting all walks of life. To deal with it, we need to make use of scientifically sound tools and models. Operations research (OR), as a well-established field which focuses on deploying analytical tools to solving decision making problems, comes to the rescue. In this paper, by examining the OR literature and practices related to pandemics (including COVID-19), we discuss what OR can help to tackle challenges under COVID-19. We classify the literature into three stages, namely "before pandemic", "during pandemic" and "after pandemic". We examine the related literature and reveal the respective research areas and OR methods employed. Then, we propose a future research agenda. Finally, we establish the sense-and-respond OR framework regarding what specific actions should be taken to cope with COVID-19 from the perspectives of governments, healthcare and non-profit-making organizations, and businesses. We believe that the findings of this paper lay the solid foundation to stimulate further OR studies to combat COVID-19.

show abstract

Section: Using Information Technologies (Rd8)mentioning

confidence: 99%

Fighting against COVID-19: what operations research can help and the sense-and-respond framework

Choi

2021

Ann Oper Res

View full text Add to dashboard Cite

show abstract

“…Reinforcement learning allows the structure to learn for an online environment, and credit assignment is based on positive and negative messages (Boden, 2018; Li et al, 2018). Reinforcement learning could be beneficial in trading securities that require a constant rebalancing of portfolios (Luo, Lin, & Zheng, 2019).…”

Section: The Techniques Of Ai Used In Finance and Financial Marketsmentioning

confidence: 99%

“…One study adopted an actor‐critic “Deep Deterministic Policy Gradient,” a kind of reinforcement learning algorithm, to examine trading on the Chinese stock exchange. It showed that this algorithm was better than a simple recurrent reinforcement learning algorithm but that both can make more money on the stock exchange than a random walk (Luo et al, 2019). The authors feel that their algorithm can help human traders make trading decisions by providing trading signals (long, short, or neutral).…”

Section: The Application Sectors Of Ai In Finance and Financial Marketsmentioning

confidence: 99%

Artificial intelligence techniques in finance and financial markets: A survey of the literature

Milana

Ashta

2021

Strategic Change

View full text Add to dashboard Cite

Both academic and nonacademic literature is evolving following the oscillating development of artificial intelligence (AI) and computing power's evolution in their application to finance and financial markets. The limits to economic growth encountered after the financial crisis and, successively, the recent pandemic bursts have posed new challenges to AI-related technologies. The surveyed publications have given hope in harvesting improved efficiency, new data, information, advisory and management services, risk mitigation, and some unanswered questions regarding negative impacts on sustainable growth and increasing economic welfare.

show abstract

“…Deepmind (Lillicrap et al, 2015) proposed an improved version of the actor-critic algorithm named DDPG, which uses the deep neural networks to estimate the optimal policy function instead of choosing the action based on a specific distribution (Qiu et al, 2019). Luo et al (2019) give a brief literature review on DDPG method. DDPG can avoid the curse of dimensionality compared with Markov decision process (Van Otterlo and Wiering, 2012) and Q-learning (Watkins and Dayan, 1992) which require the discretization of the state.…”

Section: Deep Deterministic Policy Gradientmentioning

confidence: 99%

Software-in-the-Loop Combined Reinforcement Learning Method for Dynamic Response Analysis of FOWTs

Chen

2021

Front. Mar. Sci.

View full text Add to dashboard Cite

Floating offshore wind turbines (FOWTs) still face many challenges on how to better predict the dynamic responses. Artificial intelligence (AI) brings a new solution to overcome these challenges with intelligent strategies. A new AI technology-based method, named SADA, is proposed in this paper for the prediction of dynamic responses of FOWTs. Firstly, the methodology of SADA is introduced with the selection of Key Disciplinary Parameters (KDPs). The AI module in SADA was built in a coupled aero-hydro-servo-elastic in-house program DARwind and the policy decision is provided by the machine learning algorithms deep deterministic policy gradient (DDPG). Secondly, a set of basin experimental results of a Hywind Spar-type FOWT were employed to train the AI module. SADA weights KDPs by DDPG algorithms' actor network and changes their values according to the training feedback of 6DOF motions of Hywind platform through comparing the DARwind simulation results and that of experimental data. Many other dynamic responses that cannot be measured in basin experiment could be predicted in higher accuracy with this intelligent DARwind. Finally, the case study of SADA method was conducted and the results demonstrated that the mean values of the platform's motions can be predicted by AI-based DARwind with higher accuracy, for example the maximum error of surge motion is reduced by 21%. This proposed SADA method takes advantage of numerical-experimental method and the machine learning method, which brings a new and promising solution for overcoming the handicap impeding direct use of traditional basin experimental technology in FOWTs design.

show abstract

A novel CNN-DDPG based AI-trader: Performance and roles in business operations

Cited by 54 publications

References 31 publications

Fighting against COVID-19: what operations research can help and the sense-and-respond framework

Fighting against COVID-19: what operations research can help and the sense-and-respond framework

Artificial intelligence techniques in finance and financial markets: A survey of the literature

Software-in-the-Loop Combined Reinforcement Learning Method for Dynamic Response Analysis of FOWTs

Contact Info

Product

Resources

About