A new AI algorithm augments an AI system’s capacity to identify patterns and learn on its own absent human intervention.
Researchers at the University of Technology Sydney (UTS) have created a new AI algorithm called Torque Clustering, which greatly enhances an AI system’s ability to learn and identify patterns in data on its own, without human input. Torque Clustering reveals hidden patterns by rapidly analyzing large datasets across multiple disciplines including astronomy, biology, chemistry, finance, medicine, and psychology.
“In nature, animals learn by observing, exploring, and interacting with their environment, without explicit instructions. The next wave of AI, ‘unsupervised learning’ aims to mimic this approach,” UTS Distinguished Professor CT Lin explains. “Nearly all current AI technologies rely on ‘supervised learning’, an AI training method that requires large amounts of data to be labeled by a human using predefined categories or values, so that the AI can make predictions and see relationships.” But supervised learning has several limitations. “Labeling data is costly, time-consuming, and often impractical for complex or large-scale tasks.” Torque Clustering works without labeled data, uncovering the inherent structures and patterns within datasets.
A game-changer
Rigorously tested on 1,000 diverse datasets, the Torque Clustering algorithm achieved an average adjusted mutual information (AMI) score – a measure of clustering results – of 97.7%. Compare that to state-of-the-art methods which only achieve scores in the 80% range. This type of fully autonomous machine learning could accelerate innovation in general artificial intelligence, particularly in robotics and autonomous systems. The open-source code in Torque Clustering has been made available to researchers helping to optimize movement, control, and decision-making.
Top 3 Takeaways
- University of Technology Sydney (UTS) researchers have created a new AI algorithm called Torque Clustering, facilitating an AI system’s ability to learn and identify patterns in data on its own.
- Torque Clustering reveals hidden patterns by rapidly analyzing large datasets.
- The Torque Clustering algorithm achieved an average adjusted mutual information (AMI) score of 97.7% compared to current methods which only achieve scores in the 80% range.
