Course wrap up CS 486/686 University of Waterloo Lecture 24: July - - PowerPoint PPT Presentation

▶
course wrap up
SMART_READER_LITE
LIVE PREVIEW

Course wrap up CS 486/686 University of Waterloo Lecture 24: July - - PowerPoint PPT Presentation

Oct 12, 2022 213 likes •344 views

Course wrap up CS 486/686 University of Waterloo Lecture 24: July 24, 2017 Outline Course wrap up Final exam info (see course website) Other AI courses, options and programs AI research AI jobs 2 CS486/686 Lecture

slide-1
SLIDE 1

Course wrap up

CS 486/686 University of Waterloo Lecture 24: July 24, 2017

slide-2
SLIDE 2

CS486/686 Lecture Slides (c) 2017 P. Poupart

2

Outline

  • Course wrap up
  • Final exam info (see course website)
  • Other AI courses, options and programs
  • AI research
  • AI jobs
slide-3
SLIDE 3

CS486/686 Lecture Slides (c) 2017 P. Poupart

3

Topics Covered

  • Search algorithms
  • Probabilistic Inference
  • Decision Making under Uncertainty
  • Machine Learning
  • A bit of Natural Language Processing
  • A bit of Robotics
slide-4
SLIDE 4

CS486/686 Lecture Slides (c) 2017 P. Poupart

4

Topics That We Didn’t Cover

  • Computer Vision
  • Natural Language Processing
  • Robotics
  • Multi-agent Systems
slide-5
SLIDE 5

CS486/686 Lecture Slides (c) 2017 P. Poupart

5

Other AI courses

  • CS485/685: Theoretical Machine Learning (Shai Ben-

David S18)

  • CS489/698: Intro to Machine Learning (Yaoliang Yu

F17, Pascal Poupart W18)

  • CS484/684: Computer Vision
  • CS499R: Readings in Computer science
  • CS499T: Honours thesis
slide-6
SLIDE 6

CS486/686 Lecture Slides (c) 2017 P. Poupart

6

AI Option

  • Bachelor in CS or SE with AI Option (starting F18)
  • Joint option between CS and Engineering
  • 7 courses

– CS 486: Intro to AI – CS 492: Social Implications of Computing – One of

  • CS489: Intro to Machine Learning
  • CS485: Machine Learning Theory

– One of

  • SE 380: Intro to Feedback Control
  • ECE 486: Robot Dynamics and Control
  • ECE 380: Analog Control
  • MTE 546: Multi-sensor Data Fusion

– Three additional elective courses

slide-7
SLIDE 7

CS486/686 Lecture Slides (c) 2017 P. Poupart

7

Elective Courses in AI Option

  • Three additional elective courses among

– CS489: Intro to Machine Learning – CS485: Machine Learning Theory – CS452: Real-time Systems – STAT341: Intro to Computational Statistics – STAT440: Computational Inference – STAT441: Statistical Learning: Classification – STAT444: Statistical Learning: Function estimation – ECE423: Embedded Computer Systems – ECE481: Digital Control – ECE486: Robot Dynamics and Control – ECE488: Multivariate Control – MTE544: Autonomous Robotics – MSCI446: Data Warehousing and Mining – SYDE372: Pattern Recognition – SYDE552: Machine Intelligence – SYDE556: Simulating Neurobiological Systems

slide-8
SLIDE 8

CS486/686 Lecture Slides (c) 2017 P. Poupart

8

Data Science

  • https://uwaterlo.ca/data-science
  • Bachelor’s degree in data science

– Available soon

  • Master’s degree in data science

– Joint program between CS and Statistics – 8 courses

  • STAT 845: Statistical Concepts for Data Science
  • STAT 847: Exploratory Data Analysis
  • CS 651: Data-Intensive Distributed Computing
  • One course among

– CS648 Database System Implementation – CS689 Intro to Machine Learning – CS685 Machine Learning Theory

  • 4 additional elective courses
slide-9
SLIDE 9

CS486/686 Lecture Slides (c) 2017 P. Poupart

9

Waterloo AI Institute

  • Web: uwaterloo.ca/artificial-intelligence-institute
  • Joint institute between Math and Engineering
  • Foundational AI

– Machine learning, statistical learning, data mining – Probabilistic models, knowledge discovery, knowledge representation – Intelligent agents and game theory – Optimization and decision making – Data science and analytics – Affective computing and human-machine interaction

  • Operational AI

– Scalable AI: commercialization by both small startups and large corporations – Compact AI: deployed wherever cost, energy and bandwidth are limited – Secure AI: private data protected locally, metadata shared by cloud users – Accessible AI: tailored for ease of use – Dependable AI: with reliable performance regardless of connectivity – Transparent AI: performance of safety critical systems can be certified

slide-10
SLIDE 10

CS486/686 Lecture Slides (c) 2017 P. Poupart

10

AI research group in CS

  • Web: ai.uwaterloo.ca
  • Professors:

– Peter van Beek (applied machine learning, constraint prog.) – Shai Ben David (learning theory) – Robin Cohen (multi-agent systems, user modeling) – Jesse Hoey (health informatics, applied machine learning, computer vision) – Kate Larson (game theory, mechanism design) – Edith Law (social computing, human-computer interaction) – Richard Mann (computational audio, computer vision) – Pascal Poupart (machine learning, natural language processing, health informatics) – Yaoliang Yu (machine learning, optimization)

slide-11
SLIDE 11

CS486/686 Lecture Slides (c) 2017 P. Poupart

11

Pascal’s research projects

  • Machine Learning and Planning

– Sum-Product Networks – Bayesian learning – Reinforcement learning – Data Complexity Analysis

  • Natural language processing

– Conversational agents – Natural language understanding

  • Health Informatics

– Mobile health, activity tracking, emotion recognition

slide-12
SLIDE 12

CS486/686 Lecture Slides (c) 2017 P. Poupart

12

AI jobs

  • Data Science: golden age of Machine Learning
  • AI is revolutionizing Computer Science

– Machine Learning: new paradigm that avoids programming – Computer vision: computers can finally see – Natural Language Processing: new paradigm for HCI

  • All large companies have AI R&D groups

– Google, Microsoft, Facebook, IBM, Amazon, Baidu, Huawei

  • Many small companies use AI

– ProNavigator, TalkIQ, Focal Systems, HockeyTech, Kik Interactive, In the Chat