Machine Learning Introduction incldues why is machine learning important? How Does Machine Learning Work? Types of Machine Learning, Supervised Learning, Forms of Supervised Learning, Bayesian Learning, Learning in Bayesian Networks. | Machine Learning Introduction • Why is machine learning important? – Early AI systems were brittle, learning can improve such a system’s capabilities – AI systems require some form of knowledge acquisition, learning can reduce this effort • KBS research clearly shows that producing a KBS is extremely time consuming – dozens of man-years per system is the norm • in some cases, there is too much knowledge for humans to enter (., common sense reasoning, natural language processing) – Some problems are not well understood but can be learned (., speech recognition, visual recognition) – AI systems are often placed into real-world problem solving situations • the flexibility to learn how to solve new problem instances can be invaluable – A system can improve its problem solving accuracy (and possibly efficiency) by learning how to do something better How Does Machine Learning Work? • Learning in general breaks down into one of two forms – Learning something new • no prior knowledge of the domain/concept so no previous representation of that knowledge • in ML, this requires adding new information to the knowledge base – Learning something new about something you already knew • add to the knowledge base or refine the knowledge base • modification of the previous representation – new classes, new features, new connections between them – Learning how to do something better, either more efficiently or with more accuracy • previous problem solving instance (case, chain of logic) can be “chunked” into a new rule (also called memoizing) • previous knowledge can be modified – typically this is a parameter adjustment like a weight or probability in a network that indicates that this was more or less important than previously thought Types of Machine Learning • There are many ways to implement ML – Supervised vs. Unsupervised vs. Reinforcement • is there a “teacher” that rewards/punishes right/wrong answers? – Symbolic vs. Subsymbolic vs. Evolutionary • at what level