I am joining Karen Fasimpaur for this deep learning MOOC dive starting in January and following (#dlmooc). As ever, I selfishly look forward to the deep learning that follows in her wake. Deep for me anyway.
One of the elements I hope to bring to this work is that of iconoclasm and idiosyncracy. I think that if we are going to use the ‘depth’ metaphor (and I will tell you right now that I am deeply suspicious of it) that we must extend it to what satisfies our own unique intuition of what amounts to ‘deep’. I suspect that it is all about a particular cognitive embodiment.
In other words, I want to ask and answer this question: what does deep learning feel like. If I can get a clearer sense of that, then I think the Hewlett Foundation’s deep learning principles will click into place. As they are right now, they feel smooth and weightless in my mind’s pocket. I want that deep learning to gain heft and texture as it rolls around in our connected minds.
I look … well, I was going to say ‘forward to it’, but I am reminded that the Greeks of old had a very different notion of the future. They thought of the past and present as being in front of them, the future behind them. That feels pretty deep to me if logic is an example of deep learning.
What I mean is that even if you can’t see the future, you can still feel it whistling up behind you into the present and off into the past. I trust that we will make something of this deep and windy Muse.