Let us continue this discussion in chat.

Can you please Edit your post to be more simple and design a use case that we can follow through its details.

you'll answer some question from the only lesson that you have learned (it's not even necessary for the question to be in the exam of the 100 courses, and here you fail), besides like I said, you need more gates to open to proceed to the following layers, and if you have revised only one lesson, that will not pass through the necessary activation layers ! It's not guarantee that your model will start to learn from the only course that you have indicated cause it's a blackbox, and if you so badly want your model to memorize, why bother with learning !

no, it not enough for the model to learn anything, simply imagine you're studying for an exam that have 100 courses, but you only revised 1, in your case you have learned almost nothing, and the same goes for ANN's. In technical terms you have activation functions that needs to process valid data to be activated in order to give results as output for the other layers, otherwise the informations will get lost in the blackbox of your model.

You have too few images, so you can't do that.

@ncasas I agree with you that the ANN's are like black boxes, but in you choose a fairly simple dataset, then have a few layers and neurons, this way you can track down the ANN's behavior.

@ncasas in NLP some concepts involving irony, sarcams, negations, jokes to name a few needs a lot of context information, and coding them is not straightforward. For example if you have the word "good" in sentence your model will likely to determine it as positive, and for the example "not as good as", what do you think you're model will classify it as? so you need to explicitly code some negation terms yourself, then even with a simple RNN you'll have a better reults.