RDC’s innovative study on transfer learning, published in Frontiers in Artificial Intelligence, demonstrates real-world application

Rich Data Co (RDC), an Australian company that leads the market for artificial intelligence (AI) for credit risk, announces the publication of the article “Modeling credit risk using transfer learning and domain adaptation” in the international peer-reviewed scientific journal Frontiers in Artificial Intelligence.

The article, which presents a 12-month study in collaboration with the School of Computer Science and Technology at the University of New South Wales (UNSW), successfully demonstrates how transfer learning can be used to predict the likelihood of credit risk for businesses with limited or non-existent credit history.

Limited data in the field of credit risk assessment can adversely affect micro, small and medium enterprises. An RDC study found that data from other credit risk-related domains, such as cards and personal loans, can be used to teach models for predicting credit risk in small business lending using a new concept called Progressive Shift Contribution.

With a focus on real-world applications, RDC The research also considered the possibility of explaining portable learning models and tested the approach to domain adaptation to improve model accuracy. Further research will explore methods of automating figurative learning, thus maximizing opportunities for those who are typically removed from the life cycle of credit to gain access to credit.

Ada Guan, CEO and co-founder of RDC, commented: “Promoting financial integration is at the heart of the RDC’s goal, and our joint research with UNSW is an important step in achieving that goal.”

“Responsible use of artificial intelligence can promote fair, inclusive and sustainable access to credit, especially for new sectors and small companies, where it is more difficult to predict results due to lack of data and higher risk. Techniques such as transfer training can bring significant benefits to financial services by taking information from one application, such as personal lending, and applying it to another.

“I would like to thank my colleagues and important authors of this article at RDC and UNSW, including Hendra Surianta, Ashesh Mahidadi, Michael Bane and Charles Guan.”

RDC’s commitment to rigorous industry research and innovation is central to its success. Through its research, RDC is developing mechanisms that will revise lending methods, solving industry problems in innovative ways and introducing new thinking to the arena.