Amex Uses Synthetic Data for Rare Fraud Patterns

Something we modeled even before machine learning to test 'nearness' to rare,  or even nonexistent patterns. 

Companies Beef Up AI Models with Synthetic Data  By The Wall Street Journal, July 28, 2021

Companies are building synthetic datasets when real-world data is unavailable to train artificial intelligence (AI) models to identify anomalies.

Dmitry Efimov at American Express (Amex) said researchers have spent several years researching synthetic data in order to enhance the credit-card company's AI-based fraud-detection models.  Amex is experimenting with generative adversarial networks to produce synthetic data on rare fraud patterns, which then can be applied to augment an existing dataset of fraud behaviors to improve general AI-based fraud-detection models.

Efimov said one AI model is used to generate new data, while a second model attempts to determine the data's authenticity. Efimov said early tests have demonstrated that the synthetic data improves the AI-based model's ability to identify specific types of fraud.

From The Wall Street Journal 

View Full Article - May Require Paid Subscription

How P&G and American Express Are Approaching AI

I am quoted in the Harvard Business Review about how P&G successfully used AI in the past to improve systems, including estimates of actual value.     This HBR article has just been reposted, and the complete article is for sale if you don't have a subscription .... Ask me for more about these efforts.  Much supporting information has also been posted here.   More details were also published in the Cognitive Systems Institute archives.

How P&G and American Express Are Approaching AI
By Thomas H. Davenport, Randy Bean

Published March 31, 2017

There is a tendency with any new technology to believe that it requires new management approaches, new organizational structures, and entirely new personnel. That impression is widespread with cognitive technologies — which comprises a range of approaches in artificial intelligence (AI), machine learning, and deep learning. Some have argued for the creation of “chief cognitive officer” roles, and certainly many firms are rushing to hire experts with deep learning expertise. “New and different” is the ethos of the day. .... 

Two good examples of combining well-established practices with cognitive technology to achieve business success are American Express and Procter & Gamble. Both firms are actively undertaking cognitive technology initiatives.  Both are well into their second centuries; they wouldn’t still be here if they weren’t able to accommodate change well and introduce new technology effectively. We spoke with top executives at each of these firms about the rise of cognitive in their organizations. Ash Gupta is President of Global Credit Risk and Information Management at American Express, and Guy Peri is Chief Data Officer and Vice President of Information Technology at P&G. Both executives have longstanding track records of success at their respective organizations, having seen business and technology change come and go for 20 years or more.

How it will impact business, industry, and society.

Both organizations have a considerable history with artificial intelligence. Gupta at American Express reminded us of the Authorizer’s Assistant, which was one of the more successful rule-based expert systems of the late 1980s. As described in a popular Harvard Business Review article on that generation of technology, the system made recommendations to human authorizers whether to approve large purchase transactions by cardholders.

P&G also built and employed a number of rule-based expert systems. In addition to Peri, the current CDO, we also spoke with Franz Dill, a retired P&G IT manager who focused on AI during the 80s and 90s. He said that the most well-known expert system they developed was one that blended Folgers coffee (no longer a P&G brand). This system, Dill noted, saved P&G in excess of $20 million dollars a year in green coffee costs. The company also built an expert system that helped advertisers at P&G to use, modify, and reuse the company’s advertising assets.

Both American Express and P&G are companies that have explored artificial intelligence over the years, and while the technology may have changed, the established yet innovative approaches that these firms take to incorporating new technologies and capabilities continues to evolve. Their fundamentally sound innovation practices provide a foundation for evolution. The attributes of their respective approaches to cognitive technology include .... "