Artificial Intelligence: Applications and Cognitive Science - PowerPoint PPT Presentation

1 / 14
About This Presentation
Title:

Artificial Intelligence: Applications and Cognitive Science

Description:

Artificial Intelligence: Applications and Cognitive Science Christopher G. Prince Department of Computer Science University of Minnesota Duluth – PowerPoint PPT presentation

Number of Views:717
Avg rating:3.0/5.0
Slides: 15
Provided by: dUmnEduc7
Category:

less

Transcript and Presenter's Notes

Title: Artificial Intelligence: Applications and Cognitive Science


1
Artificial IntelligenceApplications and
Cognitive Science
  • Christopher G. Prince
  • Department of Computer Science University of
    Minnesota Duluth
  • Duluth, MN USA
  • chris_at_cprince.com
  • http//www.cprince.com

2
Overview
Christopher G. Prince Verbmobil works in
domains travel planning, appointment scheduling,
and remote PC maintenance.
  • Applications of Artificial Intelligence (AI)
  • Verbmobil
  • Honda bipedal robotics
  • Cognitive Science and AI
  • RatCog
  • KidCause
  • Conclusions

3
Verbmobil
4
Honda Bipedal Robotics
5
RatCog Cognitive Modeling
  • Questions
  • How do you program a rat?
  • What programming language is required?
  • What kinds of maps can rats form?
  • Neural basis of memory?
  • Goals
  • Create computer models of rat spatial behavior
  • Construct a workbench for these models

6
RatCog Simulator
  • Methods
  • Software agents
  • Maze environment simulation
  • Plugin rat models experimenters
  • Plugins implemented
  • Stupid rat experimenter
  • Identity match rat experimenter
  • Symmetry rat experimenter

7
RatCog
8
RatCog Software Architecture
Rat Plugin Server
Environment
Graphical User Interface (GUI)
Experimenter Plugin Server
Database
9
KidCause
  • Questions
  • Can we program a robot to develop a theory?
  • Can a robot learn a theory like a child?
  • Goals
  • Improve the generalizations of AI systems
  • Build human-centered computer systems

10
KidCause Robotic Modeling
  • Method
  • Use information from child psychology
  • Apply this knowledge to robotics
  • Design developmental algorithms
  • Specific approach
  • Model language learning
  • Language is involved in theory learning

11
KidCause Model 1 Input
12
KidCause Model 1Proposed Software Architecture
Feature Detector (objects)
Learning algorithm
Video Tape Input
Feature Detector (face)
Feature Detector (speech)
13
Conclusions
  • Artificial Intelligence
  • Good progress using engineering techniques
  • May have limits
  • Cognitive Science
  • Reverse engineering biological intelligence
  • Can assist AI

14
Acknowlegements
  • Eric Mislivec, Alex Kosolapov
  • Beth Hennon, Kathy Hirsh-Pasek, Roberta Golinkoff
  • J. K. Nath, Sam Storie
Write a Comment
User Comments (0)
About PowerShow.com