Abstract: Artificial intelligence (AI) has accelerated the advancement of financial services by identifying hidden patterns from data to improve the quality of financial decisions. However, in addition to commonly desired attributes, such as model accuracy, financial services demand trustworthy AI with properties that have not been adequately realized. These properties of trustworthy AI are interpretability, fairness and inclusiveness, robustness and security, and privacy protection. Here, we review the recent progress and limitations of applying AI to various areas of financial services, including risk management, fraud detection, wealth management, personalized services, and regulatory technology. Based on these progress and limitations, we introduce FinBrain 2.0, a research framework toward trustworthy AI. We argue that we are still a long way from having a truly trustworthy AI in financial services and call for the communities of AI and financial industry to join in this effort.

金融大脑 2.0: 当金融遇到可信人工智能

[1]Aloud ME, Alkhamees N, 2021. Intelligent algorithmic trading strategy using reinforcement learning and directional change. IEEE Access, 9:114659-114671.

[2]Azzone M, Barucci E, Moncayo GG, et al., 2022. A machine learning model for lapse prediction in life insurance contracts. Expert Syst Appl, 191:116261.

[3]Baesens B, Höppner S, Verdonck T, 2021. Data engineering for fraud detection. Decis Support Syst, 150:113492.

[4]Bai MJ, Zheng Y, Shen Y, 2022. Gradient boosting survival tree with applications in credit scoring. J Oper Res Soc, 73(1):39-55.

[5]Cao SS, Yang XX, Chen C, et al., 2019. TitAnt: online real-time transaction fraud detection in Ant Financial. Proc VLDB Endow, 12(12):2082-2093.

[6]Carcillo F, Le Borgne YA, Caelen O, et al., 2021. Combining unsupervised and supervised learning in credit card fraud detection. Inform Sci, 557:317-331.

[7]Chen C, Liang C, Lin JB, et al., 2019a. InfDetect: a large scale graph-based fraud detection system for E-commerce insurance. IEEE Int Conf on Big Data, p.1765-1773.

[8]Chen C, Fu CL, Hu X, et al., 2019b. Reinforcement learning for user intent prediction in customer service bots. Proc 42^nd Int ACM SIGIR Conf on Research and Development in Information Retrieval, p.1265-1268.

[9]Chen CC, Zhou J, Wang L, et al., 2021. When homomorphic encryption marries secret sharing: secure large-scale sparse logistic regression and applications in risk control. Proc 27^th ACM SIGKDD Conf on Knowledge Discovery & Data Mining, p.2652-2662.

[10]Chen ZY, Van Khoa LD, Teoh EN, et al., 2018. Machine learning techniques for anti-money laundering (AML) solutions in suspicious transaction detection: a review. Knowl Inform Syst, 57(2):245-285.

[11]Cheng DW, Xiang S, Shang CC, et al., 2020. Spatio-temporal attention-based neural network for credit card fraud detection. Proc AAAI Conf on Artif Intell, 34(1):362-369.

[12]Cheng XQ, Liu SH, Sun XQ, et al., 2021. Combating emerging financial risks in the big data era: a perspective review. Fundam Res, 1(5):595-606.

[13]Chou YC, Chen CT, Huang SH, 2022. Modeling behavior sequence for personalized fund recommendation with graphical deep collaborative filtering. Expert Syst Appl, 192:116311.

[14]Corbett-Davies S, Pierson E, Feller A, et al., 2017. Algorithmic decision making and the cost of fairness. Proc 23^rd ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.797-806.

[15]Cui LM, Seo H, Tabar M, et al., 2020. DETERRENT: knowledge guided graph attention network for detecting healthcare misinformation. Proc 26^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.492-502.

[16]Dash S, Günlük O, Wei D, 2018. Boolean decision rules via column generation. Proc 32^nd Int Conf on Neural Information Processing Systems, p.4655-4665.

[17]Dastile X, Celik T, Potsane M, 2020. Statistical and machine learning models in credit scoring: a systematic literature survey. Appl Soft Comput, 91:106263.

[18]Djeundje VB, Crook J, Calabrese R, et al., 2021. Enhancing credit scoring with alternative data. Expert Syst Appl, 163:113766.

[19]Doering J, Kizys R, Juan AA, et al., 2019. Metaheuristics for rich portfolio optimisation and risk management: current state and future trends. Oper Res Perspect, 6:100121.

[20]Dumitrescu E, Hué S, Hurlin C, et al., 2022. Machine learning for credit scoring: improving logistic regression with non-linear decision-tree effects. Eur J Oper Res, 297(3):1178-1192.

[21]Ehrentraud J, Ocampo DG, Garzoni L, et al., 2020. Policy Responses to Fintech: a Cross-Country Overview. FSI Insights on Policy Implementation, No. 23. Bank for International Settlements.

[22]Etmann C, Lunz S, Maass P, et al., 2019. On the connection between adversarial robustness and saliency map interpretability. Proc 36^th Int Conf on Machine Learning, p.1823-1832.

[23]Fiore U, De Santis A, Perla F, et al., 2019. Using generative adversarial networks for improving classification effectiveness in credit card fraud detection. Inform Sci, 479:448-455.

[24]Floridi L, 2019. Establishing the rules for building trustworthy AI. Nat Mach Intell, 1(6):261-262.

[25]Forough J, Momtazi S, 2021. Ensemble of deep sequential models for credit card fraud detection. Appl Soft Comput, 99:106883.

[26]G20, 2019. G20 Japan: AI Principles. https://www.g20-insights.org/wp-content/uploads/2019/07/G20-Japan-AI-Principles.pdf [Accessed on Jan. 14, 2022].

[27]Gao QJ, Xu DL, 2019. An empirical study on the application of the evidential reasoning rule to decision making in financial investment. Knowl-Based Syst, 164:226-234.

[28]Ghosh Dastidar K, Jurgovsky J, Siblini W, et al., 2022. NAG: neural feature aggregation framework for credit card fraud detection. Knowl Inform Syst, 64(3):831-858.

[29]Gianini G, Fossi LG, Mio C, et al., 2020. Managing a pool of rules for credit card fraud detection by a game theory based approach. Fut Gener Comput Syst, 102:549-561.

[30]Gomes C, Jin Z, Yang HL, 2021. Insurance fraud detection with unsupervised deep learning. J Risk Insur, 88(3):591-624.

[31]Hassani H, Huang X, Silva E, et al., 2020. Deep learning and implementations in banking. Ann Data Sci, 7(3):433-446.

[32]Herland M, Bauder RA, Khoshgoftaar TM, 2019. The effects of class rarity on the evaluation of supervised healthcare fraud detection models. J Big Data, 6(1):21.

[33]Hickman E, Petrin M, 2021. Trustworthy AI and corporate governance: the EU's ethics guidelines for trustworthy artificial intelligence from a company law perspective. Eur Bus Org Law Rev, 22(4):593-625.

[34]Hu BB, Zhang ZQ, Shi C, et al., 2019. Cash-out user detection based on attributed heterogeneous information network with a hierarchical attention mechanism. Proc AAAI Conf on Artif Intell, 33(1):946-953.

[35]IEEE, 2017. EAD: a Vision for Prioritizing Human Well-Being with Autonomous and Intelligent Systems. https://standards.ieee.org/news/2017/ead_v2 [Accessed on Jan. 13, 2022].

[36]Jiang JX, Ni BY, Wang CP, 2021. Financial fraud detection on micro-credit loan scenario via Fuller location information embedding. Web Conf, p.238-246.

[37]Kazemi HR, Khalili-Damghani K, Sadi-Nezhad S, 2021. Tuning structural parameters of neural networks using genetic algorithm: a credit scoring application. Expert Syst, 38(7):e12733.

[38]Kim E, Lee J, Shin H, et al., 2019. Champion-challenger analysis for credit card fraud detection: hybrid ensemble and deep learning. Expert Syst Appl, 128:214-224.

[39]Kriebel J, Stitz L, 2022. Credit default prediction from user-generated text in peer-to-peer lending using deep learning. Eur J Oper Res, 302(1):309-323.

[40]Lappas PZ, Yannacopoulos AN, 2021. A machine learning approach combining expert knowledge with genetic algorithms in feature selection for credit risk assessment. Appl Soft Comput, 107:107391.

[41]Lee SI, Yoo SJ, 2020. Multimodal deep learning for finance: integrating and forecasting international stock markets. J Supercomput, 76(10):8294-8312.

[42]Li Q, Tan JH, Wang J, et al., 2020. A multimodal event-driven LSTM model for stock prediction using online news. IEEE Trans Knowl Data Eng, 33(10):3323-3337.

[43]Li Z, Hui PR, Zhang P, et al., 2021. What happens behind the scene? Towards fraud community detection in E-commerce from online to offline. Web Conf, p.105-113.

[44]Li ZC, Huang M, Liu GJ, et al., 2021. A hybrid method with dynamic weighted entropy for handling the problem of class imbalance with overlap in credit card fraud detection. Expert Syst Appl, 175:114750.

[45]Liang C, Liu ZQ, Liu B, et al., 2019. Uncovering insurance fraud conspiracy with network learning. Proc 42^nd Int ACM SIGIR Conf on Research and Development in Information Retrieval, p.1181-1184.

[46]Lin WL, Sun L, Zhong QW, et al., 2021. Online credit payment fraud detection via structure-aware hierarchical recurrent neural network. Proc 30^th Int Joint Conf on Artificial Intelligence, p.3670-3676.

[47]Liu WW, Guo J, Sonboli N, et al., 2019. Personalized fairness-aware re-ranking for microlending. Proc 13^th ACM Conf on Recommender Systems, p.467-471.

[48]Liu ZQ, Chen CC, Yang XX, et al., 2018. Heterogeneous graph neural networks for malicious account detection. Proc 27^th ACM Int Conf on Information and Knowledge Management, p.2077-2085.

[49]Madiega TA, 2019. EU Guidelines on Ethics in Artificial Intelligence: Context and Implementation. https://www.europarl.europa.eu/RegData/etudes/BRIE/ 2019/640163/EPRS_BRI(2019)640163_EN.pdf [Accessed on Jan. 19, 2022].

[50]Mehrabi N, Morstatter F, Saxena N, et al., 2021. A survey on bias and fairness in machine learning. ACM Comput Surv, 54(6):115.

[51]Mohassel P, Zhang YP, 2017. SecureML: a system for scalable privacy-preserving machine learning. IEEE Symp on Security and Privacy, p.19-38.

[52]Ozbayoglu AM, Gudelek MU, Sezer OB, 2020. Deep learning for financial applications: a survey. Appl Soft Comput, 93:106384.

[53]Proença HM, van Leeuwen M, 2020. Interpretable multiclass classification by MDL-based rule lists. Inform Sci, 512:1372-1393.

[54]Qi Y, Xiao J, 2018. Fintech: AI powers financial services to improve people's lives. Commun ACM, 61(11):65-69.

[55]Sawhney R, Mathur P, Mangal A, et al., 2020. Multimodal multi-task financial risk forecasting. Proc 28^th ACM Int Conf on Multimedia, p.456-465.

[56]Sobreira Leite G, Bessa Albuquerque A, Rogerio Pinheiro P, 2019. Application of technological solutions in the fight against money laundering—a systematic literature review. Appl Sci, 9(22):4800.

[57]Soui M, Gasmi I, Smiti S, et al., 2019. Rule-based credit risk assessment model using multi-objective evolutionary algorithms. Expert Syst Appl, 126:144-157.

[58]Sun SR, Wu B, Zhang ZX, et al., 2019. A hierarchical insurance recommendation framework using GraphOLAM approach. IEEE/ACM Int Conf on Advances in Social Networks Analysis and Mining, p.757-764.

[59]Suzumura T, Zhou Y, Baracaldo N, et al., 2019. Towards federated graph learning for collaborative financial crimes detection. https://arxiv.org/abs/1909.12946

[60]Théate T, Ernst D, 2021. An application of deep reinforcement learning to algorithmic trading. Expert Syst Appl, 173:114632.

[61]Wang DX, Lin JB, Cui P, et al., 2019. A semi-supervised graph attentive network for financial fraud detection. IEEE Int Conf on Data Mining, p.598-607.

[62]Wang JY, Zhang Y, Tang K, et al., 2019. AlphaStock: a buying-winners-and-selling-losers investment strategy using interpretable deep reinforcement attention networks. Proc 25^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.1900-1908.

[63]Wang L, Li PP, Xiong K, et al., 2021. Modeling heterogeneous graph network on fraud detection: a community-based framework with attention mechanism. Proc 30^th ACM Int Conf on Information & Knowledge Management, p.1959-1968.

[64]Wang Z, Zhang W, Liu N, et al., 2021. Scalable rule-based representation learning for interpretable classification. Proc 35^th Conf on Neural Information Processing Systems, p.30479-30491.

[65]Wei D, Dash S, Gao T, et al., 2019. Generalized linear rule models. Proc 36^th Int Conf on Machine Learning, p.6687-6696.

[66]Wu J, Wang C, Xiong LD, et al., 2019. Quantitative trading on stock market based on deep reinforcement learning. Int Joint Conf on Neural Networks, p.1-8.

[67]Wu ZH, Pan SR, Chen FW, et al., 2020. A comprehensive survey on graph neural networks. IEEE Trans Neur Netw Learn Syst, 32(1):4-24.

[68]Xing E, 2018. SysML: on system and algorithm co-design for practical machine learning. Proc 24^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.2880.

[69]Xu BB, Shen HW, Sun BJ, et al., 2021. Towards consumer loan fraud detection: graph neural networks with role-constrained conditional random field. Proc AAAI Conf on Artif Intell, 35(5):4537-4545.

[70]Xu H, Liu XR, Li YX, et al., 2021. To be robust or to be fair: towards fairness in adversarial training. Proc 38^th Int Conf on Machine Learning, p.11492-11501.

[71]Xu K, Fu CL, Zhang XL, et al., 2020. aDMSCN: a novel perspective for user intent prediction in customer service bots. Proc 29^th ACM Int Conf on Information & Knowledge Management, p.2853-2860.

[72]Yan C, Li YQ, Liu W, et al., 2020. An artificial bee colony-based kernel ridge regression for automobile insurance fraud identification. Neurocomputing, 393:115-125.

[73]Yang F, Qiao YN, Huang C, et al., 2021a. An automatic credit scoring strategy (ACSS) using memetic evolutionary algorithm and neural architecture search. Appl Soft Comput, 113:107871.

[74]Yang F, He K, Yang LX, et al., 2021b. Learning interpretable decision rule sets: a submodular optimization approach. Proc 34^th Conf on Neural Information Processing Systems, p.27890-27902.

[75]Yang S, Zhang ZQ, Zhou J, et al., 2021. Financial risk analysis for SMEs with graph-based supply chain mining. Proc 29^th Int Joint Conf on Artificial Intelligence, p.4661-4667.

[76]Zhang XW, Han YC, Xu W, et al., 2021. HOBA: a novel feature engineering methodology for credit card fraud detection with a deep learning architecture. Inform Sci, 557:302-316.

[77]Zhang YL, Zhou J, Zheng WH, et al., 2019. Distributed deep forest and its application to automatic detection of cash-out fraud. ACM Trans Intell Syst Technol, 10(5):55.

[78]Zheng WB, Yan L, Gou C, et al., 2021. Federated meta-learning for fraudulent credit card detection. Proc 29^th Int Joint Conf on Artificial Intelligence, p.4654-4660.

[79]Zheng XL, Zhu MY, Li QB, et al., 2019. FinBrain: when finance meets AI 2.0. Front Inform Technol Electron Eng, 20(7):914-924.

[80]Zhong QW, Liu Y, Ao X, et al., 2020. Financial defaulter detection on online credit payment via multi-view attributed heterogeneous information network. Web Conf, p.785-795.

[81]Zhou ZH, Feng J, 2017. Deep forest: towards an alternative to deep neural networks. Proc 26^th Int Joint Conf on Artificial Intelligence, p.3553-3559.

[82]Zhu XQ, Ao X, Qin ZD, et al., 2021. Intelligent financial fraud detection practices in post-pandemic era. Innovation, 2(4):100176.

[83]Zhu YC, Xi DB, Song BW, et al., 2020. Modeling users' behavior sequences with hierarchical explainable network for cross-domain fraud detection. Web Conf, p.928-938.