MACHINE LEARNING 張銘軒 譚恆力 1

OUTLINE OVERVIEW HOW DOSE THE MACHINE “ LEARN ” ? ADVANTAGE OF MACHINE LEARNING ALGORITHM TYPES  SUPERVISED LEARNING  UNSUPERVISED LEARNING  REINFORCEMENT LEARNING MACHINE LEARNING AND DATA MINING APPLICATIONS CONCLUSION REFERENCES 2

OVERVIEW Machine learning is the science of getting computers to act without being explicitly programmed. In present time, machine learning has given us self-driving cars, practical speech recognition, effective web search, and a vastly improved understanding of the human genome. Machine learning is so pervasive today that you probably use it dozens of times a day without knowing it. 3

HOW DOSE THE MACHINE “ LEARN ” ? Machine learning deals with the construction and study of algorithms that can learn from data. Such algorithms operate by building a model based on inputs and using that to make predictions or decisions, rather than following only explicitly programmed instructions. 4

ADVANTAGE OF MACHINE LEARNING The emphasis of machine learning is on automatic methods. The goal is to devise learning algorithms that do the learning automatically without human’s assistance. Rather than program the computer to solve the task directly, in machine learning, we seek methods by which the computer will come up with its own program based on examples that we provide. 5

ALGORITHM TYPES Based on the desired outcome of the algorithm or the type of input available during training of the machine, machine learning tasks can be in following several forms:  Supervised learning  Unsupervised learning  Semi-supervised learning  Reinforcement learning  … 6

SUPERVISED LEARNING Supervised learning is the machine learning task of inferring a function from labeled training data. The training data consist of a set of training examples. In supervised learning, each example is a pair consisting of an input object and a desired output value. 7

UNSUPERVISED LEARNING Unsupervised learning is closely related to the problem of density estimation in statistics. However unsupervised learning also encompasses many other techniques that seek to summarize and explain key features of the data. Approaches to unsupervised learning include:  Clustering (e.g., k-means, mixture models, hierarchical clustering)  Hidden Markov models  Blind signal separation 8

REINFORCEMENT LEARNING Reinforcement learning is an area of machine learning inspired by behaviorist psychology, concerned with how software agents should take actions in an environment so as to maximize some notion of cumulative reward. Reinforcement learning is different from standard supervised learning in that correct input/output pairs are never presented, nor sub-optimal actions explicitly corrected. 9

Machine Learning and Data Mining Machine learning is sometimes conflated with data mining, although data mining focuses more on exploratory data analysis. The two areas overlap in many ways, data mining uses many machine learning methods.  Retail: Market basket analysis, Customer Relationship Management (CRM)  Finance: Credit scoring, fraud detection  Manufacturing: Control, robotics, troubleshooting  Medicine: Medical diagnosis  Telecommunications: Spam filters, intrusion detectionWeb mining: Search engines ... 10

APPLICATIONS Nowadays, machine learning is widely used, e.g.:  Machine perception  Computer vision (including object recognition)  Syntactic pattern recognition  Search engines  Medical diagnosis  Bioinformatics  Brain-machine interfaces  Detecting credit card fraud  Stock market analysis  Classifying DNA sequences 11

CONCLUSION These days, machine learning techniques are being widely used to solve real-world problems by storing, manipulating, extracting and retrieving data from large sources. Research in machine learning has been vigorous and fruitful, and we can look forward to a continuation of good work in this area. 12

References Wikipedia Berkeley Statistical Machine Learning Journal of Machine Learning Research Michael I. Jordan “Statistics and Machine Learning” mtt?context=3 Robert E. Schapire “Foundations of Machine Learning” 13