# Is it a good idea to train a neural network to classify images without base-hypothesis?

I'm a relative beginner in deep-learning (understand by that, I'm doing my first kaggle competition right now, and I have loads to learn still) and I was just wondering something. Let's say you have pathology/biopsy tissue images from patients dying from a disease A and patients died from other causes (whatever causes actually but not related to disease A).

To date, I think we can say that nobody actually really knows what causes at the level of a biopsy the disease A. My idea, as my group could have actually a lot of these biopsies for both groups, would be to use them to fuel a neural network. Why would I do that? Biopsies images are rather complex images, and maybe some fine details are hard to guess for a human being, or maybe the sum of some details is actually important to tell whether disease A kills the patient or not. But again, I don't think anybody could come and say: on those tissue biopsy, the sign(s) for disease A are x, y, z.

My question then becomes a bit more theoretical: given the fact that you have enough data to actually give chances to the algorithm to find differences, is it a good idea to train a neural network without having actually any idea of what could differentiate the two groups? Do you know examples of such a strategy? How hard is it afterwards - in the case of a rather good accuracy - to understand what makes it so recognisable?

• It's not clear to me what you actually want to do: are you saying that you have a labelled dataset of $N$ instances where the labels are either $l_1$ (it has the disease) or $l_2$ (it does not have the disease), and your goal is to train a neural network to predict the label of an unseen image? If that's your setting, what is your question? Are you asking if you (the human) need to know what patterns in the images are associated with "diseases" in order to train the neural network? The answer is clearly no: you don't need that, but maybe you need that to understand if the net is doing well. – nbro Jul 31 '20 at 20:59
• Thanks for answering. My question was indeed: does one need to have an hypothesis to build a neural network or can one build hypothesis based on what the neural network found (if successful) – H. Vilbert Aug 1 '20 at 12:14