Embed
Email

Learning

Document Sample

Shared by: huanglianjiang1
Categories
Tags
Stats
views:
1
posted:
12/23/2011
language:
pages:
20
AI in Robotics





 Kevin Cormier





 Tyler Magalhaes





 Kristopher Schlemmer

AI in Robotics



 Topics:

 What is learning?

 Why do we care?

 What types of learning are there?

 How do we apply it?

What is Learning





 The cognitive process of acquiring

knowledge





 Learning is the acquisition and development

of memories and behaviors, including skills,

knowledge, understanding, values, and

wisdom.

Learning for a Robot



 The ability of a machine to improve its

performance based on previous results.





 Extraction of knowledge from examples





 Execution of algorithms modified during

runtime automatically.

Learning – Why it matters



 Because the world is  A robot, if it's to be

dynamic and no autonomous and

programmer has useful, must be able

experienced to learn from it's

everything, a environment,

program cannot be determine what data

written to handle is important, and

every single possible adjust it's behaviors

encounter a robot based on it's findings

may have in the real while ignoring

world. unimportant data.

Learning



 Many types

 Supervised

 Bayesian

 Unsupervised

 Reinforcement

Supervised Learning



 Uses inputs with

known outputs to

learn a function.

 Two types:

 Regression – A

continuous

function

 Classification –

Grouping of data

Supervised Learning - Benefits



 No algorithm/problem understanding

necessary

 Can reason about some things it's never

seen before (make inferences)

 Able to find complex relationships without

user interaction.

 Has teacher or training algorithm to help

ensure learning based on desired output.

Supervised Learning - Pitfalls



 If the training data doesn't contain the necessary info to learn the

function, algorithm will never converge on a solution

 A proper algorithm and structure must be chosen. Too complex a

problem and you can't learn it. Too complex an

algorithm/structure and it will take far longer than it has to to learn

everything.

 Have to have training/testing datasets

 Takes time to train

 Can be slow

 Can be over trained

 Back Propagating

Neural Network

 Feed forward training inputs

 Compare to outputs

 Slightly correct network

 Repeat

 Uses already computed datasets

 Teaching algorithm determines how much to

correct based on error and training info.

 Forced to generalize rules

Decision Trees



 Uses best classifying characteristics to build

decisions trees

 Great for classification problems

 Uses already classified data to determine

best attributes to calculate the data best.

 Teaching algorithm determines what's “best”

(speed, accuracy...)

Applications



 Classification

 Is that object too large to climb over?

 Is the ground soft or hard?

 Is that bridge safe to cross?

 Regression

 Steering

 Safe speed calculation

Real World Applications



 Path finding

 localization done used rfid tags and

scanners. Neural nets used to recognize and

approach rfid tags

 Goal Recognizing

 used for some image processing, recognizing

goals in different environments

Real World Applications





 Threat Detection





 Agent is given feature data about different objects in a

terrain path and needs to decide if they are threats to be

avoided (ex. Fire, tree, land mine...) or if they are benign

(puddle, small rock, shadow)

Bayesian Learning



 Represents variables and the probabilities

that they're independent

 Can be used for speech recognition

 Used to model knowledge, information

retrieval, and image processing

 Most others train, then put into use.

Bayesian networks can learn while in use

(learn from real world experience as well as

training data)

Unsupervised Learning



 Involves learning a pattern among presented inputs

without being provided desired outputs.

 Agent may be able to learn patterns in a given input set, but

will not have a concept of “right or wrong” without being told

examples of desired and undesired outputs.

 Used for development of probabilistic reasoning systems

(e.g. weather prediction).

 Example: autonomous combat planes can predict tactical

maneuvers of targets and provide effective

countermeasures.

Reinforcement Learning



 Very similar to supervised learning, except feedback is

provided by the environment, rather than the

developer.

 Example – combat plane is presented with different

training scenarios and asked to shoot down targets. It

would receive points for shooting enemy targets and

lose points for shooting friendly targets.

 Example – Vacuuming robot accumulates points for

vacuuming up dirt, no points for going over clean

areas, and loses points on a steady time scale (to

provide sense of urgency).

Neural Networks



 Models the brain’s cognitive system to solve problems,

make decisions, and prompt actuators.

 Series of feature data provided by sensors are presented to

the neural network. Different configurations of networks

provide different types of outputs (YES/NO decisions or

linear predictions).

 Can be done supervised (using back-propagation) or

unsupervised.

 In today’s practice, neural networks are most suitable for

classification problems. This would be useful for a combat

plane to identify friendly fire vs. valid targets.

Other AI Topics



 Inductive Learning – Make greater than first-order

conclusions. Used to map way-point plan for

achieving a goal.



 Ensemble Learning – Making integrated conclusions

by combining sensor data. Learn weighting functions.

Are all inputs required to make conclusions?



 Committees of Agents

References



http://cs.nyu.edu/~raia/docs/proposal.pdf

http://en.wikipedia.org/wiki/Unsupervised_learning

Stuart Russel, Peter Norvig - Artificial Intelligence A Modern Approach Second

Edition

Google Definitions



Related docs
Other docs by huanglianjiang...
ИТОГИ
Views: 0  |  Downloads: 0
AW Nov08 PT FINAL.indd
Views: 0  |  Downloads: 0
Michigan Arts
Views: 0  |  Downloads: 0
Educational Attainment - CT.gov Home
Views: 0  |  Downloads: 0
frankfurt_doctors_1107
Views: 8  |  Downloads: 0
Perceptionsoct07
Views: 0  |  Downloads: 0
4300 LP 4 x 2
Views: 2  |  Downloads: 0
20090515154711
Views: 0  |  Downloads: 0
CPChicago
Views: 0  |  Downloads: 0
Parent Release Form
Views: 1  |  Downloads: 0
By registering with docstoc.com you agree to our
privacy policy

You are almost ready to download!

You are almost ready to download!