Transfer learning is very simple. Very complicated generative adversarial models / deep RL are still research domains and it is not clear it's good enough yet f

Transfer learning is very simple. Very complicated generative adversarial models / deep RL are still research domains and it is not clear it's good enough yet for major commercial applications. ResNet is a very simple idea. The spelling/grammar errors are minor and not very important.