CSE 573:

Intro to Artificial Intelligence

Hanna Hajishirzi

slides adapted from

Dan Klein, Pieter Abbeel ai.berkeley.edu
And Dan Weld, Luke Zettelmoyer
 Website
o Website
o tentative schedule
o lecture slides
o course policies, etc.

o https://courses.cs.washington.edu/courses/cse573/21wi/
 Course Staff

Hanna Hajishirzi hannaneh@cs Mondays Zoom

Aida Amini amini91@cs Fridays Zoom
Josh Gardner jpgard@cs Tuesday/Thursdays Zoom

• Office hours
• Schedule on the website
• TAs: concepts, projects, homework
• Hanna: concepts, high level guidance, homework
• Introductions?
 Logistics
o Canvas: grades, submitting assignments:
o private matters – private messages
o if your message is not answered promptly enough, use the staff email:

o Ed: Discussion board: ask and answer questions; announcements

 Course Format
o Programming Assignments
o 4 projects
o Python
o Autograded
o Give you hands-on experience with the algorithms
o I expect you to get 100% on projects

o Written homeworks
o 2 written homeworks
o Gives you a more conceptual understanding of the material
 Course Format (continued)
o Paper report
o Learn how to read and criticize research papers

o Final Project:
o Encourage to pick a project related to your research
o We will provide recommendations for picking projects
o There will be a proposal day.
 Prerequisites
o Data Structure or Equivalent:
CSE 332

o Math:
o Basic exposure to probability and data structures

o Programming – Familiar with Python

o There is a 0th project (P0)
 Textbook
o Not required, but for students who want to read more
we recommend
o Russell & Norvig, AI: A Modern Approach, 3rd Ed.

o Warning: Not a course textbook, so our presentation does

not necessarily follow the presentation in the book.
 Course Policies
o Grade:
o Your grade will be: 5% class participation, 5% paper reports, 30%
programming assignments, 30% homeworks, and 30% project.
o Assignments should be done individually unless otherwise
specified.
o Late Policy: Six penalty-free late day for the whole quarter;
maximum 4 days per assignment. No late day for the final.
 Today

o What is artificial intelligence (AI)?

o What can AI do?

o What is this course?

AI
 What is AI?
The science of making machines that:

Think like Think rationally

people

Act like people Act rationally

 Rational Decisions
We’ll use the term rational in a very specific, technical way:
▪ Rational: maximally achieving pre-defined goals
▪ Rationality only concerns what decisions are made
(not the thought process behind them)
▪ Goals are expressed in terms of the utility of outcomes
▪ Being rational means maximizing your expected utility

A better title for this course would be:

Computational Rationality
Maximize Your
Expected Utility
Maximize Your
Expected Utility
Maximize Your
Expected Utility
Maximize Your
Expected Utility
 What About the Brain?
▪ Brains (human minds) are very good
at making rational decisions, but not
perfect
▪ Brains aren’t as modular as software,
so hard to reverse engineer!
▪ “Brains are to intelligence as wings
are to flight”
▪ Lessons learned from the brain:
memory and simulation are key to
decision making
 Designing Rational Agents

o An agent is an entity that perceives and acts.

o A rational agent selects actions that maximize its
(expected) utility.
o Characteristics of the percepts, environment, and
action space dictate techniques for selecting rational
actions
o This course is about:
o General AI techniques for a variety of problem types

Environment
o Learning to recognize when and how a new problem Sensors
can be solved with an existing technique Percepts

Agent
?
Actuators Actions
 Topics in This Course
o Part I: Intelligence from Computation
o Fast search
o Adversarial and uncertain search
o Part II: Reasoning under Uncertainty
o Bayes’ nets
o Decision theory
o Machine learning
o Throughout: Applications
o Natural language, vision, robotics, games, …
 AI
Machine Learning
[learning decisions;
sometimes independent]

Robots Rational Human-AI

[physically
Agents Interaction
embodied]
[decisions]

NLP
Computer
Vision
 Today

o Course overview

o What is artificial intelligence (AI)?

o What can AI do?

o What is this course?

A (Short) History of AI
 A (Short) History of AI
o 1940-1950: Early days
o 1943: McCulloch & Pitts: Boolean circuit model of brain
o 1950: Turing's “Computing Machinery and Intelligence”

o 1950—70: Excitement: Look, Ma, no hands!

o 1950s: Early AI programs, including Samuel's checkers program, Newell &
Simon's Logic Theorist, Gelernter's Geometry Engine
o 1956: Dartmouth meeting: “Artificial Intelligence” adopted
o 1965: Robinson's complete algorithm for logical reasoning

o 1970—90: Knowledge-based approaches

o 1969—79: Early development of knowledge-based systems
o 1980—88: Expert systems industry booms
o 1988—93: Expert systems industry busts: “AI Winter”

o 1990—2012: Statistical approaches

o Resurgence of probability, focus on uncertainty
o General increase in technical depth
o Agents and learning systems… “AI Spring”?
o 2012— present: Excitement: Look, Ma, no hands!
o Big Data, big compute, neural networks
o Some re-unification of subfields
o AI is being used in industry.
 What Can AI Do?
Quiz: Which of the following can be done at present?

o Play a decent game of Jeopardy?

o Win against any human at chess?
o Win against the best humans at Go?
o Play a decent game of tennis?
o Grab a particular cup and put it on a shelf?
o Unload any dishwasher in any home?
o Drive safely along the highway?
o Drive safely along University Avenue?
o Buy a week's worth of groceries on the web?
o Buy a week's worth of groceries at QFC?
o Discover and prove a new mathematical theorem?
o Perform a surgical operation?
o Unload a known dishwasher in collaboration with a person?
o Translate spoken Chinese into spoken English in real time?
o Write an intentionally funny story?
 Unintentionally Funny Stories
o One day Joe Bear was hungry. He asked his friend
Irving Bird where some honey was. Irving told him
there was a beehive in the oak tree. Joe walked to
the oak tree. He ate the beehive. The End.

o Henry Squirrel was thirsty. He walked over to the

river bank where his good friend Bill Bird was sitting.
Henry slipped and fell in the river. Gravity drowned.
The End.

o Once upon a time there was a dishonest fox and a vain crow. One day the crow was
sitting in his tree, holding a piece of cheese in his mouth. He noticed that he was holding
the piece of cheese. He became hungry, and swallowed the cheese. The fox walked over
to the crow. The End.
[Shank, Tale-Spin System, 1984]
 Natural Language
o Speech technologies (e.g. Siri)
o Automatic speech recognition (ASR)
o Text-to-speech synthesis (TTS)
o Dialog systems

o Language processing technologies

o Question answering
o Machine translation

o Web search
o Text classification, spam filtering, etc…
 Computer Vision

• Object Recognition
• Scene Classification
• Image Segmentation
• Human Activity Recognition
Scene
Segmentation

Object Recognition
https://pjreddie.com/darknet/yolo/
 Google Goggles

Smile Detection
Leaf Snap
 Image captioning: What begins to work
The flower was so
vivid and attractive.

Blue flowers are running

We sometimes do well: 1 out of 4 times, machine
rampant in my garden.
captions were preferred over the original Flickr captions:

Spring in a white dress.

Blue flowers have no scent. Small white
flowers have no idea what they are.

Scenes around the lake on my bike ride.

This horse walking along the road as we drove by.
 But many challenges remain
(better examples of when things go awry)

The couch is definitely bigger than it Yellow ball suspended in water.

looks in this photo.

My cat laying in my duffel bag.

A high chair in the trees.
 Game Agents
o Classic Moment: May, '97: Deep Blue vs. Kasparov
o First match won against world champion
o “Intelligent creative” play
o 200 million board positions per second
o Humans understood 99.9 of Deep Blue's moves
o Can do about the same now with a PC cluster

o 1996: Kasparov Beats Deep Blue

“I could feel --- I could smell --- a new kind of intelligence across the table.”

o 1997: Deep Blue Beats Kasparov

“Deep Blue hasn't proven anything.”

Text from Bart Selman, image from IBM’s Deep Blue pages
 Game Agents
o Reinforcement learning

Pong Enduro Beamrider Q*bert

 2016

AlphaGo deep RL defeats Lee Sedol (4-1)

 Simulated Agents

[Schulman, Moritz, Levine, Jordan, Abbeel, ICLR 2016]

 Robotics
o Robotics
o Part mech. eng.
o Part AI
o Reality much
harder than
simulations!

o Technologies
o Vehicles
o Rescue
o Help in the home
o Lots of automation…

o In this class:
o We ignore mechanical aspects
o Methods for planning
o Methods for control

Images from UC Berkeley, Boston Dynamics, RoboCup, Google

Robots
 Robocup
o https://www.youtube.com/watch?v=_PC-V5GJP6Q
Robocup
Tools for Predictions & Decisions
 Decision Making

• Applied AI in many kinds of automation:

• Scheduling, airline routing
• Route planning
• Medical diagnosis
• Web search
• Spam classification
• Automated help desks
• Smarter devices, like cameras
• Fraud detection
• Product recommendation
• … Lots more!
 Today

o Course overview

o What is artificial intelligence (AI)?

o What can AI do?

o What is this course?

 Designing Rational Agents
o An agent is an entity that perceives and acts.
o A rational agent selects actions that maximize its
(expected) utility.
o Characteristics of the percepts, environment, and
action space dictate techniques for selecting rational
actions
o This course is about:
o General AI techniques for a variety of problem types

Environment
o Learning to recognize when and how a new problem Sensors
can be solved with an existing technique Percepts

Agent
?
Actuators Actions
Maximize Your
Expected Utility
 Topics in This Course
o Part I: Intelligence from Computation
o Fast search
o Adversarial and uncertain search
o Part II: Reasoning under Uncertainty
o Bayes’ nets
o Decision theory
o Machine learning
o Throughout: Applications
o Natural language, vision, robotics, games, …
 Pac-Man as an Agent

Agent Environment
Sensors
Percepts
?
Actuators Actions

Pac-Man is a registered trademark of Namco-Bandai Games, used here for educational purposes
 Assignments: Pac-man

Originally developed at UC Berkeley:

http://www-inst.eecs.berkeley.edu/~cs188/pacman/pacman.html
 This course vs. others

Agent

Environment
Sensors
o CSE 515 – Stat methods Percepts
o CSE 517 – NLP
o CSE 546 – ML ?

o CSE 571 – Robotics

o CSE 576,7 – Vision Actuators
Actions
o Advanced RL
 PS1: Search

Goal:
• Help Pac-man find his way
through the maze

Techniques:
• Search: breadth-first, depth-first,
etc.
• Heuristic Search: Best-first, A*,
etc.
 PS2: Game Playing
Goal: Techniques:
• Play Pac-man! • Adversarial Search: minimax,
alpha-beta, expectimax, etc.
 PS3: Ghostbusters
Goal:
• Help Pac-man hunt down the
ghosts

Techniques:
• Probabilistic models: HMMS,
Bayes Nets
• Inference: State estimation and
particle filtering
 PS4: Reinforcement Learning

Goal: Techniques:
• Help Pac-man learn • Planning: MDPs, Value Iterations
about the world • Learning: Reinforcement Learning
 Important This Week
• Important this week:
• Check out canvas--- our main resource for assignments and grades
• Check out website – for schedule and slides
• Check out Ed – for discussions; we are going to add everyone to Ed
• P0: Python tutorial is out

• Also important:
• Office Hours start next week.