This article has been reviewed according to Science X's editorial process and policies. Editors have highlighted the following attributes while ensuring the content's credibility: Artificial intelligence programs used to classify images could be trained much more cheaply using a surprisingly simple method inspired by the childhood game of 20 Questions, according to new research posted to the arXiv preprint server by a team from the University of Bristol. The research, presented Wednesday, Sept. 16, at the Allerton Conference on Communication, Control and Computing in Illinois, U.S., shows how simple binary classifiers, each trained in a matter of minutes on a standard laptop, can be combined using a string of yes-or-no questions to perform complex classification tasks.
These tasks normally require tens of thousands of graphics processing units to train, costing millions of dollars. Sidharth Jaggi, professor of mathematics at the University of Bristol School of Mathematics, explained, "If you were out on a walk and wanted to identify an unusual species of plant you spotted using an AI app, a current classifier program would likely be designed to carefully separate millions, if not billions, of different types of objects from each other. "In this new piece of work, we are able to mathematically prove—and back up by empirical validation—that even if we ask 'simple random questions,' the answers can be combined to perform complex tasks.
"The key observation is that no complex coordination of the simple binary classifiers is required. There just need to be enough of them, and that number is surprisingly small. The implication is that our algorithms have a far lower computational cost, are more robust and are easier to deploy at scale—all properties that are critical for real-world use." The approach works especially well for AI systems that operate across multiple devices or directly within smart devices, such as sensors, robots and edge devices that process data close to where it is generated.
Because each question is answered independently, the overall system can still produce reliable results even if some individual answers are incorrect. Ioannis Papageorgiou, who carried out the research while working as a senior research associate at the University of Bristol, said, "What is exciting about this approach is that a very large and difficult classification problem can be broken down into lots of much simpler yes-or-no decisions, chosen at random. Each individual classifier only needs to answer one of these simple questions, but together they can identify from millions of possibilities.
"As artificial intelligence becomes increasingly embedded in our lives, from health care and transportation to manufacturing and national infrastructure, the challenge is shifting. The key question is no longer just how to make AI systems more powerful, but how to make them efficient, trustworthy and resilient in real-world conditions." The team's work forms part of the Informed AI research hub at the University of Bristol, which tackles a range of foundational problems spanning mathematics, information theory and AI safety. By providing these theoretical foundations for efficiency, robustness and reliability, Informed AI research aims to ensure that future U.K.
AI systems are not only innovative but also safe, trustworthy and socially deployable. As AI continues to move into everyday environments, these principles will be essential for maintaining public confidence and delivering long-term value for the economy and society at large. Ioannis Papageorgiou et al, Fundamental limits of distributed multiclass classification from simple binary decisions, arXiv (2026).
Extract — continue reading at the source.
