Heuristic (Informed)
Search
(Where we try to be smarter in
how we choose among
alternatives)
R&N III: Chapter 3.5
R&N II: Chap. 4, Sect. 4.13
1
Search Algorithm
1. INSERT(initial-node,FRINGE)
2. Repeat:
a. If empty(FRINGE) then return failure
b. n REMOVE(
Support Vector Machines
CSE537 Artificial Intelligence, Fall 2016, Stony Brook University
Instructor: Heeyoung Kwon ([email protected])
Date: 13 Oct 2016
Slides Courtesy: Minh Hoai Nguyen
Linear classifiers Which line is better?
2
Linear classifie
Informed Search
Niranjan Balasubramanian
Fall 2016
September 13th 2016
Informed Search
Motto
Use information about how close we are to the goal.
g(A)
A
S
h(A)
G
g(B)
B
h(B)
Costs
g(A) Cost to reach node A from the start node S. i.e., cost of a specific
p
Question Answering Part 2
Semantic Parsing
Markov Logic Networks
Niranjan Balasubramanian
Stony Brook University
Slides based on material from:
Jonathan Berant, Google
Tushar Khot, AI2
Outline
Question Answering
Types of QA tasks and approaches
Macro-Rea
Natural Language Processing:
Language Modeling
Niranjan Balasubramanian
CS537 AI
Oct 25, 2016
Natural Language
Mechanism for communicating thoughts, ideas, emotions, and more.
What is NLP?
Building natural language interfaces to computers (devices more ge
Local Search
AI CS 537 Fall 2016
Niranjan Balasubramanian
Sep 20th 2016
Credits:
AIMA Slides, and Mausam
Outline
What is local search?
What kind of problems are local search techniques used for?
Three types of local search techniques
Hill climbing and it
Reasoning Under Uncertainty Part 1
Niranjan Balasubramanian
Sep 29th 2016
CS 537 Artificial Intelligence
Credits
AIMA Book
Many slides directly from:
Dan Weld and Dieter Fox
(and recursively, from all those credited on their slides)
Uncertainty
Agents mus
Search Algorithms
CSE 537 Artificial Intelligence Fall 2016
Niranjan Balasubramanian
Stony Brook University
Outline
What is search?
How to formulate search problems?
Types of Search Algorithms
Uninformed Search
Breadth first
Uniform cost
Depth-first
Syntax
Model Theory
Proof Theory and Resolution
Propositional Logic
C. R. Ramakrishnan
CSE 537
1
Syntax
2
Model Theory
3
Proof Theory and Resolution
Compiled at 14:31 on 2016/11/01
AI
Propositional Logic
CSE 537
1 / 22
Syntax
Model Theory
Proof Theory and
Question Answering
Niranjan Balasubramanian
Many Slides from:
Sanda Harabagiu, Tao Yang
Chris Manning, David Ferrucci,
Watson Team, and Paul Fodor
Outline
Question Answering
Types of QA tasks and approaches
Macro-Reading-based QA
[Today]
A General Archi
Learning from Examples
Niranjan Balasubramanian
Fall 2016 CS 537
Credits:
AIMA Book
Slides on Nave Bayes from:
Dan Klein and Pieter Abbeel.
Outline
Introduction
Nave Bayes Classification
Logistic Regression
Next class
Support Vector Machines
Unsuper
First-Order Logic (FOL)
Slides by:
Tuomas Sandholm
Carnegie Mellon University
Computer Science Department
Propositional Logic is weak
Hard to identify individuals (e.g: Mary)
Cant directly talk about properties of individuals
(e.g: Mary is tall)
No Gen
Graph Coloring using CSP and Min-conflicts Local Search
Assignment 1
Due: 11:59 pm Nov 4th, 2016
Goals: The main goal of this assignment is to implement 1) Depth first w/ backtracking (DFSB)
and improvements to it (DFSB+), and 2) learn the details of a lo
Iterative Deepening and IDA*
Alan Mackworth
UBC CS 322 Search 6
January 21, 2013
Textbook 3.7.3
Lecture Overview
Recap from last week
Iterative Deepening
Slide 2
Search with Costs
Sometimes there are costs associated with arcs.
Def.: The cost of a path
Search Algorithm
Heuristic (Informed)
Search
1. INSERT(initial-node,FRINGE)
2. Repeat:
a. If empty(FRINGE) then return failure
b. n REMOVE(FRINGE)
c. s STATE(n)
d. If GOAL?(s) then return path or goal state
e. For every state s in SUCCESSORS(s)
i. Create
Artificial Intelligence CSE 537
Assignment 2 (Oct. 14th 2015)
Due date and time: Oct 26th 4:00pm
Submit in class.
* points max
1. Which of the following are correct?
A.
False |= True.
B.
True |= False.
C.
(AB)|=(A B).
D.
A B |= A B .
E.
A B|= AB.
F. (AB)
Artificial Intelligence CSE 537
Assignment 3 (Nov. 12th 2014)
Due date and time: Nov 25th 2:30pm (in class)
Submit in class (hardcopy).
103 points max
(Chapters covered: )
1. (pg.506 Exercise 13.3 [ 4x3 =12 points])
For each of the following statements,
Artificial Intelligence CSE 537
Assignment 1 (Sept. 24th 2015)
Due date and time: Oct. 12th in class.
Submit in class (hardcopy).
* 80 points max
Question 1: Warm-up (1 points)
What is your preferred e-mail contact?
Question 2: Intelligent Agents (9 point
Artificial Intelligence CSE 537
Assignment 4 (Dec. 2nd 2015)
Due date and time: Dec. 9th 2:30pm
Submit in class (hardcopy), use blackboard, or e-mail me ([email protected]).
* points max
1. pg. 763 Excersize18.8 [10 points]Decision Trees
Consider the f
Assignment 3: Text Classification
Due: December 2nd, 2016
Naive Bayes, Logistic Regression, Support Vector Machines, and Random Forests
This is an analysis assignment, where you will investigate the utility of different learning
algorithms for a text clas
Graph Coloring using CSP and Min-conflicts Local Search
Assignment 1
Due: 11:59 pm Nov 4th, 2016
Goals: The main goal of this assignment is to implement 1) Depth first w/ backtracking (DFSB)
and improvements to it (DFSB+), and 2) learn the details of a lo