Google Researchers: Computers Can Learn Like Humans

Google's researchers have developed a method to train artificial intelligence computers to learn like a human.

"When a new task is introduced, new adaptations overwrite the knowledge that the neural network had previously acquired," the DeepMind Google team said on its blog.

"This phenomenon is known in cognitive science as 'catastrophic forgetting,' and is considered one of the fundamental limitations of neural networks."

Google's researchers say the ability to learn tasks in a sequential fashion, or in a linear way by following a set of steps, is "crucial to the development of artificial intelligence." They taught a computer to remember old tasks by "selectively slowing down learning on the weights important for those tasks."

Their work was published February in the Proceedings of the National Academy of Sciences of the United States of America.

Google continues to make moves on the AI front. In November, Google's Neural Machine Translation system taught itself a new language. The company, like other tech giants Facebook and Microsoft, have remade themselves around aggressively pursuing AI to work in their favor.

• Scientists Use Artificial Intelligence to Catch 'Criminals' by Reading Minds
• Facebook Turns to Artificial Intelligence to Tackle Suicides