New AI Algorithm Could Lead to Truly Autonomous AI

Andrew Corselli

Researchers have developed a new AI algorithm, called Torque Clustering, that is much closer to natural intelligence than current methods. It significantly improves how AI systems learn and uncover patterns in data independently, without human guidance.

Torque Clustering can efficiently and autonomously analyze vast amounts of data in fields such as biology, chemistry, astronomy, psychology, finance and medicine, revealing new insights such as detecting disease patterns, uncovering fraud, or understanding behavior.

“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,” said Distinguished Professor CT Lin from the University of Technology Sydney (UTS).

“Nearly all current AI technologies rely on ‘supervised learning’, an AI training method that requires large amounts of data to be labelled by a human using predefined categories or values, so that the AI can make predictions and see relationships.

“Supervised learning has a number of limitations. Labelling data is costly, time-consuming and often impractical for complex or large-scale tasks. Unsupervised learning, by contrast, works without labelled data, uncovering the inherent structures and patterns within datasets.”

A paper detailing the Torque Clustering method, Autonomous clustering by fast find of mass and distance peaks, has just been published in IEEE Transactions on Pattern Analysis and Machine Intelligence, a leading journal in the field of artificial intelligence.

The Torque Clustering algorithm outperforms traditional unsupervised learning methods, offering a potential paradigm shift. It is fully autonomous, parameter-free, and can process large datasets with exceptional computational efficiency.

It has been rigorously tested on 1,000 diverse datasets, achieving an average adjusted mutual information (AMI) score — a measure of clustering results — of 97.7 percent. In comparison, other state-of-the-art methods only achieve scores in the 80 percent range.

“What sets Torque Clustering apart is its foundation in the physical concept of torque, enabling it to identify clusters autonomously and adapt seamlessly to diverse data types, with varying shapes, densities, and noise degrees,” said First Author Jie Yang, Ph.D.

“It was inspired by the torque balance in gravitational interactions when galaxies merge. It is based on two natural properties of the universe: mass and distance. This connection to physics adds a fundamental layer of scientific significance to the method.

“Last year’s Nobel Prize in physics was awarded for foundational discoveries that enable supervised machine learning with artificial neural networks. Unsupervised machine learning — inspired by the principle of torque — has the potential to make a similar impact,” said Yang.

Torque Clustering could support the development of general artificial intelligence, particularly in robotics and autonomous systems, by helping to optimise movement, control and decision-making. It is set to redefine the landscape of unsupervised learning, paving the way for truly autonomous AI. The open-source code has been made available to researchers.

Here is an exclusive Tech Briefs interview, edited for length and clarity, with Yang and Lin.

Tech Briefs: What was the biggest technical challenge you faced while developing Torque Clustering?

Yang: Developing a fully autonomous clustering method required addressing several interconnected challenges, including computational efficiency, scalability, and ensuring robustness across diverse datasets. Finding the right balance between performance and generalizability was a key aspect of the process.

Tech Briefs: What was the catalyst for this project?

Lin: The motivation stemmed from the broader goal of enhancing unsupervised learning capabilities for truly autonomous AI. We explored ways to enable clustering methods to function independently without relying on manually tuned parameters, which is crucial for real-world applications.

Tech Briefs: Can you explain in simple terms how it works please?

Yang: The algorithm leverages intrinsic structural properties within the data to identify meaningful patterns in a completely unsupervised manner. By efficiently detecting cluster formations based on specific mathematical principles and inspired by the physical concept of torque, it can automatically determine the number of clusters without external guidance.

Tech Briefs: Do you have any plans for further research/work/etc.?

Yang: There are certainly many interesting directions to explore autonomous machine learning for autonomous AI. Some ongoing work involves refining the method to further improve its adaptability to various types of data and integrating it with other machine learning techniques to expand its application scope.

Tech Briefs: If not, what are your next steps?

Lin: Currently, the focus is on ensuring broad accessibility of the method and encouraging further exploration by the research community. Future steps will be determined based on how the method is adopted and extended in different areas.

Tech Briefs: Is there anything else you’d like to add that I didn’t touch upon?

Lin: I think the key takeaway is that unsupervised learning remains an evolving field with great potential in the AI era. While significant progress has been made, there are always new challenges and opportunities that arise as we push the boundaries of machine learning.

Tech Briefs: Do you have any advice for researchers aiming to bring their ideas to fruition?

Lin: Persistence is key. Research is a continuous learning process, and overcoming setbacks is part of the journey. Staying open to new ideas, seeking constructive feedback, and iteratively refining your approach are all essential aspects of making meaningful contributions.