Understanding Spending Behavior: Recurrent Neural Network Explanation and Interpretation

Maree, Charl; Omlin, Christian W.

doi:10.1109/cifer52523.2022.9776210

Cited by 5 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is consistent with the high affinity of conscientious spenders towards residential mortgages [18]. It is compelling to expand the notion of personality traits from spending to wealth creation, i.e., to base personal investment advice on historical spending behaviour [19,20].…”

Section: Related Worksupporting

confidence: 62%

Can Interpretable Reinforcement Learning Manage Prosperity Your Way?

Maree

Omlin

2022

Self Cite

View full text Add to dashboard Cite

Personalisation of products and services is fast becoming the driver of success in banking and commerce. Machine learning holds the promise of gaining a deeper understanding of and tailoring to customers’ needs and preferences. Whereas traditional solutions to financial decision problems frequently rely on model assumptions, reinforcement learning is able to exploit large amounts of data to improve customer modelling and decision-making in complex financial environments with fewer assumptions. Model explainability and interpretability present challenges from a regulatory perspective which demands transparency for acceptance; they also offer the opportunity for improved insight into and understanding of customers. Post-hoc approaches are typically used for explaining pretrained reinforcement learning models. Based on our previous modeling of customer spending behaviour, we adapt our recent reinforcement learning algorithm that intrinsically characterizes desirable behaviours and we transition to the problem of prosperity management. We train inherently interpretable reinforcement learning agents to give investment advice that is aligned with prototype financial personality traits which are combined to make a final recommendation. We observe that the trained agents’ advice adheres to their intended characteristics, they learn the value of compound growth, and, without any explicit reference, the notion of risk as well as improved policy convergence.

show abstract

Section: Related Worksupporting

confidence: 62%

Can Interpretable Reinforcement Learning Manage Prosperity Your Way?

Maree

Omlin

2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…This hierarchical clustering provides a means of micro-segmenting customers according to their spending behavior in time. We then explained these behavioral trajectories by reproducing them using a linear regression model, and we interpreted them through locating a number of attractors that govern the dynamics of the state space [14]. We located these attractors by mapping the RNN output space into the state space through inverse regression.…”

Section: Methodsmentioning

confidence: 99%

“…Recurrent neural networks (RNNs) are able to extract this salient information when predicting personality traits from financial transactions [3]. In, [14], we gained an understanding of these extracted features by interpreting the dynamics of the RNN state space through a set of attractors. Understanding model behavior is crucial in industries such as personal finance [10].…”

Section: Background and Related Workmentioning

confidence: 99%

Reinforcement Learning with Intrinsic Affinity for Personalized Prosperity Management

Maree¹,

Omlin²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

The common purpose of applying reinforcement learning (RL) to asset management is the maximization of profit. The extrinsic reward function used to learn an optimal strategy typically does not take into account any other preferences or constraints. We have developed a regularization method that ensures that strategies have global intrinsic affinities, i.e., different personalities may have preferences for certain assets which may change over time. We capitalize on these intrinsic policy affinities to make our RL model inherently interpretable. We demonstrate how RL agents can be trained to orchestrate such individual policies for particular personality profiles and still achieve high returns.

show abstract

“…Recurrent neural networks (RNNs) [50,51] are a popular and well-known DL technique. RNN is frequently used in NLP and speech-processing applications.…”

Section: Approachmentioning

confidence: 99%

“…While there are several studies on trading strategy, price prediction, and portfolio management, there is a notable lack of research on market simulation, stock selection, hedging strategy, and risk management [62]. By utilizing DL approaches [50][51][52][53][54][55][56][57][58][59][60][61][62][63], several studies showed improvements in financial analyses, risk management, and the creation of intelligent financial systems.…”

Section: Approachmentioning

confidence: 99%

The Artificial Intelligence Revolution in Digital Finance in Saudi Arabia: A Comprehensive Review and Proposed Framework

Al-Baity

2023

Sustainability

View full text Add to dashboard Cite

Artificial Intelligence (AI) has proliferated in the last few years due to the vast data we pro-duce daily and available computing power. AI can be applied in many different sectors, such as transportation, education, healthcare, banking, and finance, among many others. The financial industry is rapidly embracing AI due to its potential for high-cost savings in financial services. AI could transform the financial sector by creating opportunities for tailored, faster, and more cost-effective services. Saudi Arabia is emerging as a fast-growing market in this industry with a strong commitment to technology-driven institutions. While AI is gaining prominence and receiving government support, it has not yet become a critical component for enhancing the efficiency of financial transactions. Limited published research on AI adoption in the Saudi Arabian financial industry calls for a comprehensive literature review to examine the current state of AI implementation in this sector. Therefore, this study explores the benefits, limitations, and challenges of leveraging AI in finance, highlighting the importance of ethical and regulatory considerations for successful AI adoption in the sector. This study’s findings reveal that research has been conducted on how AI improves processes in the financial sector by integrating critical components and efficient algorithms tailored to the industry’s needs. Based on these findings, this study proposes a sequential framework at the macro and micro levels of management to guide AI’s development and integration into the financial sector. Additionally, the framework draws insights from the existing literature to provide a detailed understanding of opportunities, challenges, and areas for improvement to maximize AI’s potential in the Saudi Arabian financial sector.

show abstract

Understanding Spending Behavior: Recurrent Neural Network Explanation and Interpretation

Cited by 5 publications

References 21 publications

Can Interpretable Reinforcement Learning Manage Prosperity Your Way?

Can Interpretable Reinforcement Learning Manage Prosperity Your Way?

Reinforcement Learning with Intrinsic Affinity for Personalized Prosperity Management

The Artificial Intelligence Revolution in Digital Finance in Saudi Arabia: A Comprehensive Review and Proposed Framework

Contact Info

Product

Resources

About