1) your input should be so that you describe your entire environment. This could be done by 8 (length)* 8 (height)* 3 (either empty space, opponent chip or your chip) = 192 input neurons. you can just import a image of the current boardstate (which is width pixels * height pixels input neurons), but this means you task the neural network with learning to ...
A robust ML model is one that captures patterns that generalize well in the face of the kinds of small changes that humans expect to see in the real world.
A robust model is one that generalizes well from a training set to a test or validation set, but the term also gets used to refer to models that generalize well to, e.g. changes in the lighting of a ...
Yes you can, a few years ago I made a simple CNN for a single Arabic phoneme classification. You can use spectogram or using MFCC / MFSC as features, as long all data has the same size (use padding or cropping if needed).
You may need RNN if you want to combine some phonemes to recognize a single word or longer.
I believe you may want to use a Sum Product Network for this task. SPNs are the state-of-the-art approach for face completion, and there are several more recent papers on this topic since the original above.
Importantly, the SPN paper also covers other approaches that work well for this task. If lower-resolution results are acceptable for your task, PCA ...
Initial state: initial position of the monkey.
climb on the crate,
get down the crate,
move the crate from one spot to another,
stack one crate on another,
walk from one spot to another,
grab bananas (if standing on the crate)
Goal test: did the monkey get the bananas?
Cost function: the number of actions completed