AI Enhances Decision Accuracy by Mimicking Humans

Researchers at Georgia Tech have developed a neural network, RTNet, that mimics human decision-making by incorporating uncertainty and evidence accumulation. This model, tested on handwritten digits, shows similar accuracy, response time, and confidence patterns to humans, potentially leading to more reliable AI systems and reducing cognitive load in daily decision-making.