Machine Learning: Talk at Pillai's college

One of the major concerns of artificial intelligence has been how to get the computer to learn things the way human beings do. This would make it simpler to program them, not having to hardcode every piece of information and where/when/how to use them for every problem. The system could learn by experience or observing the environment. However, this is a very hard problem in general. There are many types of learning, and one can classify them from different perspectives such as the purpose of learning (new knowledge, performance, exploration, etc), the nature of knowledge (procedures, rules, constraints, etc), the process of learning (induction, deduction, etc), and so on. This talk will cover the topic of machine learning broadly and then discuss 3-4 major approaches in some detail. This is likely to include inductive learning of rules, neural network based learning, evolutionary learning and statistical models. We will also introduce a nice, powerful open source tool called Weka which offers support for a variety of machine learning approaches. Students/teachers can use this to implement and experiment with machine learning problems.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License