• Atlanta
• Boston
• Dallas
• Ft. Lauderdale
• London
• Minneapolis
• New York
• Oakland
• Philadelphia
• Washington, DC

by Jennifer Gibbs

In the AI context, there are areas in which conventional supervised learning approaches are incapable of solving -- specifically in areas that encompass a large or increasing number of uncountable classes. “This is where Zero-Shot learning takes the wheel and attempts to solve these otherwise seemingly impossible tasks for a conventional supervised learning model to handle.”

Zero-shot learning (ZSL) is a machine learning pattern where an AI model is trained to recognize and categorize objects or concepts without having been given any examples of those categories or concepts in advance. The concept behind ZSL is to train a machine to mimic the way humans can naturally find similarities between data classes.

The main goal of ZSL is to gain the ability to predict the results without any training samples. ZSL is programmed to learn intermediate semantic layers and properties, then apply it to predict a new class of unseen data.

Unlike humans who already possess ZSL ability, machines typically require input labeled data to learn and then be able to adapt to variances that may naturally occur. More and more researchers are interested in automatic attribute recognition (ZSL) to compensate for areas in which there is a lack of available data. 

ZSL can unlock new levels of AI flexibility, enabling models to extend to entirely new data and tasks without additional labeling or additional training. “This allows efficiently scaling AI to new products, geographical markets, customer segments, and business needs as they emerge.” ZSL models have the potential to dynamically recognize an open-ended set of new concepts over time from descriptions alone. This adaptable intelligence (ZSL) can assist companies to cost-effectively innovate, personalize offerings, assess risks, flag anomalies, and create a “future-proof” AI that aligns with rapidly-changing business environments.

______________________________________

The opinions expressed are those of the authors and do not necessarily reflect the views of the firm or its clients. This article is for general information purposes and is not intended to be and should not be taken as legal advice.