CPSC 340: Machine Learning and Data Mining, Lecture notes of Machine Learning

Download CPSC 340: Machine Learning and Data Mining and more Lecture notes Machine Learning in PDF only on Docsity!

CPSC 340:

Machine Learning and Data Mining

Mark Schmidt

University of British Columbia

Fall 2015

Outline

1. Intro to Machine Learning and Data Mining:

• Big data phenomenon and types of data.
• Definitions of data mining and machine learning.
• Applications and impact.

2. Course Administrivia

3. Course Overview

Some images from this lecture are taken from Google Image Search.

Big Data Phenomenon

• What do you do with all this data?
  • Too much data to search through it manually.
• But there is valuable information in the data.
  • How can we use it for fun, profit, and/or the greater good?
• Data mining and machine learning are key tools we use to make sense of large datasets.

Data Mining

• Automatically extract useful knowledge from large datasets.
• Usually, to help with human decision making.

Data Mining vs. Machine Learning

• DM and ML are very similar:
  • Data mining often viewed as closer to databases.
  • Machine learning often viewed as closer AI.
• Both similar to statistics, but less emphasis on ‘correct’ models and more on computation.

Databases Data Mining MachineLearning Intelligence^ Artificial Humans in loop. Complexity of Tasks

Applications

• Spam filtering:
• Credit card fraud detection:
• Product recommendation:

Applications

• Face detection:
• Object detection:
• Sports analytics:

Applications

• Personal Assistants:
• Medical imaging:
• Self-driving cars:

Applications

• Inceptionism, mimicking art styles:

• Summary:
  • There is a lot you can do with a bit of statistics and a lot data.
• But, you should not use these methods blindly:
  • The future may not be like the past.
  • Associations do not imply causality.

Location, Dates, Webpage

• Course homepage:
  • www.cs.ubc.ca/~schmidtm/Courses/340-F
• Office hours:
  • Thursdays in ICCS 146 from 3-4.
  • Or by appointment.
• Tutorials:
  • Mondays in DMP 201 from 11-12, 2-3, and 4-5.
  • Only on weeks when assignments are due.
• Teaching Assistants:
  • Issam Laradji.
  • Sharan Vaswani.
  • Tian Qi (Ricky) Chen.
  • Yan Zhao.

CPSC 540 and Auditting 340

• There is also a graduate ML course, CPSC 540:
  • Higher workload.
  • More advanced material.
  • More implementation/theory, fewer applications.
• Auditing CPSC 340 or 540, an excellent option:
  • Pass/fail on transcript rather than grade.
  • Do 2 assignments or write a 2-page report on one technique from or attend > 90% of classes.
  • But please do this officially:
    • http://students.ubc.ca/enrolment/courses/academic- planning/audit

Assignments

• 6 Assignments worth 25% of final grade:
  • Written portion and Matlab programming.
  • Due at the start of Friday class:
    • September 18 (Friday of next week), October 2, October 16, November 6, November 20, December 4.
  • You can have up to 3 total ‘late classes’:
    • Handing in an assignment on Monday counts as one.
    • Handing in on Wednesday counts as two.
    • After that, you will get a mark of 0 for late assignments.
    • Examples:
      • you can hand in A1, A4, and A5 one day late.
      • you can hand in A2 two days late and A4 one day late.
      • you can hand in A1 three days late, and all others on time.

Getting Help

• Tutorials on Mondays before assignments due.
• Piazza for assignment/course questions:
  • piazza.com/ubc.ca/winterterm12015/cpsc
• If you do not have access to Matlab:
  • Ask for a CS guest account.
  • Purchase Matlab through the bookstore or online.
  • Use the free alternative Octave.
• You can work in groups and use any source, but:
  • Hand in your own homework.
  • Acknowledge all sources, including other students.