Development in the State of the Art of Machine Learning
Machine learning can be defined as a program that based on experience E with respect to some class of tasks T and a performance measure P improves its performance at task T, as measured by P, with experience E (Mitchell, 1997). Machine learning systems are not directly programmed to solve a problem, instead they develop based on examples of how they should behave and from trial-and-error experience trying to solve the problem.
The field of machine learning addresses the question of “how can we build computer systems which can automatically improve with experience, and what are the fundamental laws that govern all learning processes?” (Mitchell, 2006).
Different techniques are used in different subfields of machine learning, such as neural networks(Chauvin & Rumelhart, 1995), instance-based learning (Aha, Kibler & Albert, 1991), decision tree learning (Quinlan, 1993), computational learning theory (Kearns & Vazirani, 1994), genetic algorithms(Mitchell, 1996), statistical learning methods (Bishop, 1996), and reinforcement learning(Kaelbling, Littman & Moore, 1996). In recent years, machine learning has shed light on many other disciplines, such as economics, finance, and management. Many machine learning tools, including neural networks and genetic algorithms, are used in intelligent financial investing research.