Learning to forget — a weapon in the arsenal against harmful AI
With the AI summit well underway, researchers are keen to raise the very real problem associated with the technology — teaching it how to forget.
Society is now abuzz with modern AI and its exceptional capabilities; we are constantly reminded its potential benefits, across so many areas, permeating practically all facets of our lives — but also its dangers.
In an emerging field of research, scientists are highlighting an important weapon in our arsenal towards mitigating the risks of AI — ‘machine unlearning’. They are helping to figure out new ways of making AI models known as Deep Neural Networks (DNNs) forget data which poses a risk to society.
The problem is re-training AI programmes to ‘forget’ data is a very expensive and an arduous task. Modern DNNs such as those based on ‘Large Language Models’ (like ChatGPT, Bard, etc.) require massive resources to be trained — and take weeks or months to do so. They also require tens of Gigawatt-hours of energy for every training programme, some research estimating as much energy as to power thousands on households for one year.
Machine Unlearning is a burgeoning field of research that could remove troublesome data from DNNs quickly, cheaply and using less resources. The goal is to do so while continuing to ensure high accuracy. Computer Science experts at the University of Warwick, in collaboration with Google DeepMind, are at the forefront of this research.
Professor Peter Triantafillou, Department of Computer Science, University of Warwick, recently co-authored a publication ‘Towards Unbounded Machine Unlearning’. He said: “DNNs are extremely complex structures, composed of up to trillions of parameters. Often, we lack a solid understanding of exactly how and why they achieve their goals. Given their complexity, and the complexity and size of the datasets they are trained on, DNNs may be harmful to society.
“DNNs may be harmful, for example, by being trained on data with biases — thus propagating negative stereotypes. The data might reflect existing prejudices, stereotypes and faulty societal assumptions — such as a bias that doctors are male, nurses female — or even racial prejudices. More