Artificial Intelligence notes

Comments

• Might want to mention a simple voice chatbot that work across telephones uses (government departments probably use it) in the "Even more Chatterbots!" section. May be too much of a distraction given that it uses voice recognition as well.
• explain where the material above has oversimplified things, and if there are any well-known concepts or techniques that have been deliberately left out because they are too complex for this age group. This may include things that require advanced maths, advanced programming, or things where students have seen the problem but not a thorough solution
• Any of the following not mentioned above could be mentioned here: machine learning, natural language processing, singularity, commonsense knowledge, machine perception (vision, speech recognition), biologically inspired algorithms.
• What about genetic algorithm music? http://en.wikipedia.org/wiki/Evolutionary_music it seems like a good way to demonstrate a leaning program. I can’t seem to find a site that has something that we can use though I know that they exist.
• There is also simple data classification which learns using the test sample. Things like Nearest centroid and nearest neighbour. These are both simple to do and demostrates the tradeoff between speed and accuray. (NC is faster but less accurate O(1), while NN is more accurate but slower O(n^2)) http://en.wikipedia.org/wiki/Nearest_neighbor_search Kuroush has some good slides on this. Maybe we can use them.