A new study from MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) demonstrates a novel method for training AI models that significantly reduces computational costs and energy consumption. The technique, called ‘Liquid Neural Networks,’ allows models to be trained on smaller, specialized datasets and then efficiently scaled up, rather than requiring massive datasets from the start. This approach not only cuts training time and expense but also shows promise for creating more adaptable and efficient AI systems suitable for deployment on devices with limited processing power, such as smartphones or embedded sensors. The research highlights a potential shift in AI development toward greater efficiency and accessibility. Read the full article at https://technologyreview.com/2024/03/14/liquid-neural-networks-reduce-ai-training-costs/.