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 ‘Lazy Learning,’ focuses on selectively training only the most critical parts of a neural network for a given task, rather than the entire model. Researchers found this approach could cut training costs by up to 80% while maintaining comparable model performance. The method shows particular promise for large language models and other complex architectures where full-scale training is prohibitively expensive. This advancement could lower barriers to AI development and reduce the environmental impact of the industry. Read the full article at: https://technologyreview.com/2024/05/15/lazy-learning-ai-training.