Académique Documents
Professionnel Documents
Culture Documents
introducing
MACHINE LEARNING
By Mr. Amit Kenny Lecturer MSc.(IT) abk001@chowgules.ac.in
Lecture Outline
Machinelearningis programming computers to optimize a performance criterion using example data or past experience.
There is no need to learn to calculate payroll, simple interest, percentage, etc. Learningis used when: Human expertise does not exist (navigating on Mars), Humans are unable to explain their expertise (speech recognition) Solution changes in time (routing on a computer network) Solution needs to be adapted to particular cases (user biometrics)
Learning : to gain knowledge or understanding or skill by study, experience and training or modification of behavior by experience
Machine Learns when it changes structure, program or data based on its inputs or response to external information in a way that improves future performance. For Ex.: Speech Recognition Systems
Artificial Intelligence : Science and engineering of making intelligent machines, especially intelligent computer programs
Course Outline
Syllabus, References
Introduction to ML, Applications, Relation to AI Problems, Problem Spaces & Search Heuristic Search Techniques
Course Outline
Knowledge Representation using Predicate Logic Representing Knowledge using Rules Symbolic Reasoning under uncertainty Statistical Reasoning Concept Learning & General-to-Specific Ordering Decision Tree Learning Artificial Neural Networks
References Artificial Intelligence Russel & Norvig Artificial Intelligence - Rich & Knight
Course Outline
Introduction to Data Mining Tan Steibach & Kumar Many other books, sites, journals which will be intimated on a case by case basis
Unlocking ML
Arthur Samuel (1959) ML Definition: Field of Study that gives computers the ability to learn without being explicitly programmed
Computer Playing with Itself
Unlocking ML
Began Learning moves that would result in Wins & Losses Computer Played better than Arthur Samuel Himself
Tom Mitchell (1998) ML Definition: A Computer Program is said to learn from experience E with respect to a task T with performance measure P, if its performance on T as measured by P improves with
ML in an AI system
Unlocking ML
ML : give computers the ability to learn from data Present Data form sensors, internet, experiments Expect computer to make sensible decisions Traditionally categorized as : - Supervised Learning : Regression, Classification - Unsupervised Learning : Clustering - Reinforcement Learning Data Mining : Learning From Huge Databases Thinking Like the Human Brain : Neural
Unlocking ML
Applications : Game Playing : Deep Blue Autonomous Planning : NASAs remote agent program to plan and guide operations from high level goals supplied from ground Medical diagnosis Robotics Manufacturing Finance and Stocks Natural Language Processing
Unlocking ML
Learn Explore
Your Effort : Learn and Explore Try stuff out!! Machine Learning Projects
Free Time
Beautiful !
Questions
What is Machine Learning? Find Real World Applications of
ML