Researchers Identify Property Train Machines To Learn More Like Humans Do

MIT researchers have discovered that a training technique called adversarial training can help computer vision models learn more perceptually straight representations, like humans do. This property helps the models make more accurate predictions and could improve the safety of autonomous vehicles. The team also found that the task one trains a model to perform affects its perceptual straightness, with models trained to perform abstract tasks learning more perceptually straight representations than those trained to perform more fine-grained tasks.