[Knowledge-based AI] {ud409} Lesson 2: 02 - Introduction to CS7637

Optional Readings :

    1. Putting Online Learning and Learning Sciences Together
      https://www.youtube.com/watch?v=N56ghCGmWWQ
    2. Understanding the Natural and Artificial Worldshttp://courses.washington.edu/thesisd/documents/Kun_Herbert%20Simon_Sciences_of_the_Artificial.pdf

Introduction to Computational Psychometrics

Putting Online Learning and Learning Sciences Together: https://www.youtube.com/watch?v=N56ghCGmWWQ

Raven‘s Progressive Matrices

2x1 Raven’s Progressive Matrix

2x2 Raven’s Progressive Matrix

3x3 Raven’s Progressive Matrix I

Exercise: What is intelligence?

Q: if we succeed designing an AI system that can take an intelligence test, is the AI intelligent?

A from David: No, they are just processing signals and inputs in the correct way.

That‘s an interesting answer. Do humans do anything fundamentally different from processing inputs and selecting actions?

Principles of CS7637

we just dont reason from data, we also use data to pull out knowledge from memory, then we use this knowledge to generate expectations to make sense of the world.

Readings

Winston, P. (1993). Artificial Intelligence (3rd ed.). Addision-Wesley.

Stefik, M. (1995). Knowledge Systems. Morgan Kauffman: San Fransisco.

Rich, E., & Knight, K. (1991). Artificial intelligence. McGraw-Hill, New York.

Russell, S. & Norvig, P. (1995). Artificial Intelligence: A modern approach. Prentice-Hall: Englewood Cliffs.

Understanding the Natural and Artificial Worldshttp://courses.washington.edu/thesisd/documents/Kun_Herbert%20Simon_Sciences_of_the_Artificial.pdf

The Cognitive Connection

interestingly, people with autism perform about as well on the Raven‘s Test of Intelligence as neurotypical people.

These people often do not perform as well as others on other intelligent tests.

Raven‘s test consists only of visual analogy problems, while other tests include verbal tests

原文地址:https://www.cnblogs.com/ecoflex/p/10978863.html

时间: 2024-08-28 05:49:40

[Knowledge-based AI] {ud409} Lesson 2: 02 - Introduction to CS7637的相关文章

[Knowledge-based AI] {ud409} Lesson 24: 24 - Meta-Reasoning

Mistakes in Reasoning and Learning Beyond Mistakes: Knowledge Gaps The Blurred Line Between Cognition and Metacognition Strategy Selection Strategy Integration Optional Reading: Metacognition in Computation: A selected research review by Michael T. C

[Knowledge-based AI] {ud409} Lesson 26: 26 - Wrap-Up

represent => reason => represent =>.... data => knowledge => data > ...... Current Research 原文地址:https://www.cnblogs.com/ecoflex/p/10987521.html

[Knowledge-based AI] {ud409} Lesson 12: 12 - Logic

Why do we need formal logic? Inferences About Birds Winston Chapter 13: Page 284 Exercise: Inferences About Foos Predicates Conjunctions and Disjunctions Implies Notation Equivalency Truth Tables 原文地址:https://www.cnblogs.com/ecoflex/p/10982993.html

[Knowledge-based AI] {ud409} Lesson 13: 13 - Planning

Block Problem Revisited Painting a Ceiling States Optional Reading: Winston Chapter 15, pages 323-336 Operators Planning and State Spaces plan: Planning Partial Planning Detecting Conflicts Open Preconditions Hierarchical Task Network Planning Hierar

[Knowledge-based AI] {ud409} Lesson 19: 19 - Version Spaces

Optional reading: Winston Chapter 20 Incremental Concept Learning Revisited Abstract Version Spaces Visualizing Version Spaces converge model is thre representation of the world Example Food Allergies I Identification Trees decision tree learning Opt

[Knowledge-based AI] {ud409} Lesson 21: 21 - Configuration

Exercise: Designing a Basement Defining Configuration The Configuration Process left top: task right bottom: solution task v1 --- Configuration Space ---> solution v1 (search solution) solution v1 takes too long solution v1 --- Configuration Space --

[Knowledge-based AI] {ud409} Lesson 22: 22 - Diagnosis

Optional reading: Stefik, Chapter 9 diagnose as classification and abduction Exercise: Diagnosing Illness Defining Diagnosis Problems with Diagnosis as Classification Deduction, Induction, Abduction Optional Reading: Stefik, Diagnosis and Troubleshoo

[Knowledge-based AI] {ud409} Lesson 23: 23 - Learning by Correcting Mistakes

Optional Readings : Winston Chapter 18 http://courses.csail.mit.edu/6.034f/ai3/rest.pdf Questions for Correcting Mistakes Visualizing Error Detection Error Detection Algorithm Explanation-Free Repair Explaining the Mistake Correcting the Mistake Conn

PRML 02 Introduction:贝叶斯概率

引言 概率密度 期望和协方差 Expectations and covariances 1加权平均值 2 多变量权重 3 条件期望 4 函数方差 5 协方差 Bayesian Probability 5高斯分布 重回多项式拟合 1理解误差函数 2 理解规则化 贝叶斯曲线拟合 主要讲解了贝叶斯概率与统计派概率的不同.概率论,决策论,信息论(probability theory, decision theory, and information theory)是以后用到的三个重要工具,本节主要介绍概