Verseon Presents Two AI Innovations at the 2025 IEEE Conference on Artificial Intelligence

One Verseon paper covers a new method to compensate for missing or imperfect data and another covers Verseon's AutoESSV method for combining separate AI models to enhance predictive accuracy.

FREMONT, Calif., May 9, 2025 /PRNewswire/ -- Verseon has presented two separate papers at the 2025 IEEE Conference on Artificial Intelligence. One paper describes advances that preserve AI predictive accuracy when faced with missing or otherwise imperfect data. The other paper describes a method to choose ensemble strategies that combine a set of AI models using a split validation data set. IEEE will make the full text of Verseon's papers publicly available in approximately one month.

Real-life training data can often have errors or missing elements. Complete elimination of such samples can lead to undesirable reductions in training dataset size. Verseon's innovation in data imputation as presented in the first paper allows for better use of such imperfect samples without impacting the AI model's predictive accuracy.

As an example from the life sciences, estimating biological age using typically available datasets shows the power of Verseon's innovations in working with imperfect data. AI models utilize biomarker data and medical questionnaires to predict biological age, which can be a useful measure of a person's health status. These data sources are often incomplete. Verseon's new technique to handle incomplete datasets yields a 22% lower error rate when compared to best prior benchmarks for biological age prediction.

The second paper describes AutoESSV, a core machine-learning modeling technology improvement. Building on other advances described in a previously published paper from Verseon, AutoESSV is a better way to combine separate AI models. Verseon's new approach dynamically selects among multiple different ways to combine disparate AI models for optimal performance. When testing this new methodology on sixteen classification and regression datasets, Verseon's AI technology achieved a 25% reduction in correlation error for regression and a 12% reduction in classification errors relative to the state-of-the-art AutoSklearn framework.

"Increasing the accuracy of AI models is crucial to improving the real-word utility of AI. This is especially true in scenarios where already scarce data is imperfect or where different models within an AI ensemble provide better results depending on the region of the dataset," said Verseon's Head of Machine Learning Ed Ratner. "The innovations described in these papers continue to enhance the already robust capabilities of Verseon's novel AI technology."

About Verseon
Verseon International Corporation (www.verseon.com) is a clinical-stage, technology-driven pharmaceutical company transforming the delay, prevention, and treatment of disease. Using its Deep Quantum Modeling + AI platform, Verseon is rolling out a steady stream of life-changing medicines. Each of the company's drug programs features multiple novel candidates with unique therapeutic properties. None of these candidates can be found by other current methods. Verseon's fast-growing pipeline addresses major human diseases in the areas of cardiometabolic disorders and cancers. The company's supporters and advisors include multiple Nobel laureates, former heads of R&D of major pharmaceutical companies, and various key opinion leaders in medicine.

Media Contact

Walter Jones

[email protected]

(510) 225-9000

SOURCE Verseon International Corporation