With all the computing power and the exa-bytes of data at our disposal, we are still nowhere near creating a machine that can be remotely called ``cultured',' or a machine that is at least "aware of " and has the "flexible knowledge" of a human.
I will outline some of the challenges, and what needs to be done from a both mathematical modeling perspective and from the perspective of data and computation. I believe that as we listen to all the wonderful talks at the various workshops, and soak in all the great academic and experimental results, we should also keep in mind the chasm between what we aspire to do and what needs to be done.
I expect this session to be highly interactive and would encourage all the participants to collectively think about critical questions, such as: How "deep" is the current paradigm of Deep Learning?
Can we formulate a restricted version of the Turing test by the end of this program? What can we aim for in the next five years that is challenging and yet doable? What are the barriers? etc.
Back to Culture Analytics Tutorials