Stratyfy Introduces UnBias to Advance Fairness in Financial Services

By Erin Cahill.
Updated Feb 28, 2022

NEW YORK (PRWEB) February 28, 2022 - Stratyfy introduces UnBias, a standalone solution to help financial institutions and fintechs uncover, understand, and undo bias in complex financial decisions that impact the lives of millions.

Since 2019, Stratyfys patented bias mitigation has been core to its credit risk assessment and fraud detection offerings. Recently, new regulations and market forces have generated an unprecedented sense of urgency in the need to address bias, but adoption of adequate tools remains a challenge. To meet increased demand, for the first time Stratyfy is making UnBias available to customers as a standalone offering.

We are seeing lots of peers enter the space with well-intentioned but unproven bias solutions, said Laura Kornhauser, Co-Founder and CEO. But mitigating bias isnt an easy thing to do its ingrained in societys systems and data. Thats why were proud of the work weve done over the past several years to help our customers proactively tackle bias, and were excited to make it easier than ever before for them to do so.

UnBias is part of Stratyfys growing suite of transparent machine learning tools designed to help institutions automate and optimize complex decisions to reach more customers, minimize bias, and drive risk-adjusted growth. The product allows financial institutions to continuously monitor, pinpoint, and address sources of bias according to their specific needs and modeling approaches. Delivered via API, it is designed for any complex decisioning process where bias mitigation is paramount.

Bias mitigation is core to our products and to the very fabric of our company, said Kornhauser. Our no-code technology is flexible enough to complement customers current systems, simple enough to implement quickly, and powerful enough to see results in weeks.

In recent engagements, Stratyfys proprietary probabilistic rules engine (PRE) and bias correction layer have been proven to reduce bias by 3X with minimal impact on model accuracy, as measured by Area Under the Curve (AUC). Additionally, Stratyfys technology is currently being studied by FinRegLab and Stanford University in groundbreaking research to advance explainability and fairness of Machine Learning in credit underwriting. Final results of that study are expected to be published later this year.

UnBias also works in parallel with Stratyfys embedded risk assessment, which is designed to save teams the time and resources typically required for model compliance.

Bias is a bit of a hot topic in AI right now, said Kornhauser. But the truth is, at Stratyfy weve been thinking about itand acting on itsince our inception. We know that AI done right has enormous potential to expand access to financial services for so many underserved communities by giving institutions the tools to make better decisions. And through real explainability and transparency, were proving it can be done right.

About Stratyfy

Stratyfy delivers proprietary machine learning solutions for financial institutions, automating credit risk assessment, fraud detection, and other complex tasks without introducing new operational or regulatory risks. Stratyfys UnBias also helps users proactively identify and remove hidden bias in data and models. With Stratyfys transparent and interpretable solutions, institutions can seamlessly combine the precision of their data with the wisdom of their people to make better, faster decisions based on true risk and serve more customers.

Follow Stratyfy on Twitter and LinkedIn. For more information, visit http://www.stratyfy.com. For media or other inquiries, please email luke@stratyfy.com.