Ques- Neural style transfer is trained as a supervised learning task in which the goal is to input two images (x), and train a network to output a new, synthesized image (y).
It is one of the questions in the quiz of week 4. According to me, the answer should be true, as we input 2 images, style image, and content image, and get a generated image as the output. Also, it is a supervised learning task, as it is guided by the cost function. I don’t know where I am thinking incorrectly, but the answer provided is false. Can anyone help me out with this?
P.S. - I found this on the web, while looking for some help on the web, GitHub - divya-nk/deeplearning-specialization: These are my submissions to Coursera's Deep Learning Specialization Course. Includes solutions to quizzes and programming assignment. It consists of the all the answers which is against the Coursera code of honour.
You raised a very interesting question. Let me share my thoughts about this… I agree with the points shared here about this question. It is an application or a tool you can create by taking advantage of how machine learning algorithms “learn”, but we do not have a ground truth for the output as you have in supervised learning.
Regarding that repository violating the Code of Honour, many thanks for reporting this. We will escalate the issue.
Thanks a lot, @arosacastillo for your reply. I have finally understood why we classified this as an example of Unsupervised machine learning and not the other way around. Also, can you please ask the technical team to re-introduce the feature of feedback in the quizzes? It would be a great help!
You are welcome
Yes, I will pass the request to the team. @Mubsi
Btw, the issue with the repository has been already reported to Coursera. Many thanks again. Students like you help keeping a healthy community of DL learners
I am glad that I could be of some help to the community