A team of researchers from the New York University has developed an algorithm which has the ability to mimic the learning abilities of humans. This unique algorithm can view and draw concepts indistinguishable from humans.

According to study authors, their findings mark a significant development in the field that has always focused on designing outstanding robots and algorithms that can easily learn and adapt things very much like humans.

The researchers developed a ‘Bayesian Program Learning’ (BPL) framework that can shorten the learning process and make robot learning more similar to that of humans.

Brenden Lake, the paper's lead author, said, “Our results show that by reverse engineering how people think about a problem, we can develop better algorithms. Moreover, this work points to promising methods to narrow the gap for other machine learning tasks”.

The researchers in an explanation said that when an individual comes to know about a new concept like a new dance move or a new number, they just need to be exposed to only few examples of it so that they understand it and identify it in new situations.

On the other hand, a machine can carry out a number of pattern-replication tasks. They require thousands of examples in order to perform the task with the same level of accuracy as humans.

Study co-author Ruslan Salakhutdinov said that it has been quite tough for them to develop a machine that requires very less data as humans when they are learning a new concept.