Creating neural networks without human intervention

Too Long; Didn't Read

Deep Learning involves the development of deep neural networks, inspired by the way biological brains process information. To make any neural network good in its job, it first has to be trained. The answer is a technique called Neural Architecture Search (NAS) The NAS eliminates the need of humans from the process of deep learning. In a paper published in the summer of 2017, Google described how their NAS algorithm was able to outperform their human data scientists in creating highly accurate deep. NASNet is going down the path that could lower the barrier to the access of NAS algorithms.