Situation decomposition method extracts partial data which contains some rules.
Hiroshi Yamakawa
(FUJITSU LABORATORIES LTD.)
Abstract In an infantile development process, the fundamental knowledge about the external world is
acquired through learning without clear purposes. An adult is considered to use that fundamental knowledge for various works. The acquisition of the internal model in these early stages may exist as a background of the flexible high order function of human brain. However, research of such learning technology is not progressing to nowadays.
The system can improves prediction ability and reusability in the lasting work by using the result of learning without clear purposes. Then, we proposed the situation decomposition technology which chooses the partial information which emphasizes the relation "another attribute value will also change if one attribute value changes."
Situation decomposition technology is the technology of performing attribute selection and case selection simultaneously from the data structure from which each example constitutes an attribute vector. The newly introduced Matchability criteria are the amount of evaluations which becomes large, when the explanation range of the selected partial information becomes large and a strong relation exists in the inside. Processing of situation decomposition extracts plural partial situations (result of attribute selection and case selection) of corresponding to the local maximum points over this evaluation.
Furthermore, extraction of partial problem space (based on the Markov decision process) is possible using the technology which extended situation decomposition in the direction of time. In action decision task, such as robot control, partial problem space can be assigned as each module of multi-module architecture. Then, it can be efficiently adapted to unknown problem space by combining extracted plural partial problem space.
My strategy for Brain-like processing
Brain has very flexible learning ability.
The intelligent processes which has more flexible learning abilities are more close to real brain processes.
I want introduce learning ability to my system as possible.
Contents
1. Development and Autonomous Learning
2. SOIS (Self-organizing Information Selection) as Pre-task Learning
3. Delivering Matchable Principle
4. Situation Decomposition using Matchability Criterion
5. Application of Situation Decomposition
6. Conclusions & Future works
Autonomous Learning(Framework)
Outline of this talk
Pre-task learning
Self-organizingInformation Selection
Situation decomposition
Task learning
Cognitive Development
Situation Decomposition using Matchability Criterion
Matchable Principle
Matchability Criterion
Development and Autonomous Learning
Two aspects of Development
“Acquired environmental knowledge without particular goals which helps for problem solving for particular goals”→ “Pre-task Learning” in Autonomous Learning
“Calculation process which increases the predictable and/or operable object in the world”→ Enhancing prediction ability
Autonomous Learning: AL Two phases learning (Research in RWC)
Task learningExisting Knowledge
Acquiring environmental
knowledge
General factFor design
Acquiring solution for goal
goal
No reachingover the wall
Acquiringmovable paths
Generatingpath to the goal
Environment is given
Goal is given
Pre-task learning
DevelopmentToday’s Topic
Pre-task Learning helps Task Learning Autonomous Learning (AL)
Pre-task Learning Acquiring environmental knowledge without particular goal.
Task Learning Environmental knowledge speed up aacquiring solution for
goal.
In human: Adult people can solve given task quickly using enviro
nmental knowledge acquired for other goal or without particular goal.
Development ~ Pre-task Learning
DevelopmentToday’s Topic
Research topics for AL
Pre-task Learning (How to acquire environmental knowledge) Situation Decomposition using Matchability criterion
Situation Decomposition is kind of a Self-organizing Information Selection technology.
Task learning (How to use environmental knowledge) CITTA (Cognition based Intelligent Transaction Architecture)
Multi-module architecture which can combining environmental knowledge acquired during Pre-task learning
Cognitive Distance Learning Goal driven problem solver for each environmental knowledge.
DevelopmentToday’s Topic
Overview of Approaching for AL
CITTACombining environmental
knowledge
SituationDecomposition
Acquiring environmental
knowledge
Cognitive Distance Learning
Problem solver for each environmental
knowledge
Architecture
Learningalgorithm
Pre-task Learning Task Learning
SOIS (Self-organizing Information Selection) as Pre-task Learning
SOIS: Self-organizing Information Selection
Process: Selecting plural partial information from data. → “Situation Decomposition”
Criterion: Evaluation for each partial information. → Matchability Criterion
Knowledge = Set of structure.Partial Information = One kind of structure
※ SOIS could be a kind of knowledge acquiring process in development
Situation Decomposition is kind of SOISFor situation decomposition
Partial Information = Situation
Extracting plural situations which are combination of selected attributes and cases from spread sheet.
MS4
attributes
Cas
es
MS1
MS2
MS3
Delivering Matchable Principle
Two aspects of Development
“Acquired environmental knowledge without particular goals which helps solving problem for particular goals”→ “Pre-task Learning” in Autonomous Learning
“Calculation process which increases the predictable and/or operable object in the world”→ Enhancing prediction ability
How to enhance prediction ability
We needs Criterion for selecting situation.We wants to extract local structures.
Multiplex local structure is mixed in
real world data
MS4
MS1MS2
MS3
Situation Decomposition
Deriving Matchable Principal
What is Criterion for each selecting situation.
Matchable principle“Structures where a matching opportunity is
large are selected.”
Extracting structure (knowledge) without
particular goals.
Prediction is based on matching a case with experiences.
Factors in Matchable Principle To increase matching opportunity
Simplicityof Structure
Ockham’s razorMDL 、 AIC
Consistencyfor Data
Coveragefor Data
Our proposedMatchability criterion
Relation in Structure
AccuracyMinimize error
Case-increasingAttribute-increasing
Association rule
SD (Situation Decomposition ) andImplementation
Situation Decomposition
Extracting plural situations which are combination of selected attributes and cases from spread sheet.
Matchability=This criteria evaluates matching opportunity
Matchable Situation = Local maximums of Matchability
MS4
attributes
Cas
es
MS1
MS2
MS3
Formalization: Whole situation and Partial situations
Whole situation J=(D, N) : Contains N attributes and D cases.Attribute selection vector:
d = (d 1 , d 2 ,…,dD)
Case selection vector : n = (n1, n2,…,nN) Vector element di,ni are binary indicator of
selection/unselection.
Number of selected attributes: dNumber of selected cases : n
Situation decomposition extracts some matchable situations from whole situation J=(D, N) which potentially contains 2D+N partial situation.
Case selection using Segment spaceSegment space is multiplication of separation of each
selected attributes. (example: two dimension)n : Number of selected cases
Sd : Number of total
segments
rd : Number of
selected segments
※ Cases inside the chosen segments are surely chosen.
Sd =s1 s2
attribute1
attr
ibut
e2
dd SC
rC
NCNSrnM logloglog),,,( 321 dd
[Number of selected cases] n →Make Larger
[Number of total segments] Sd →Make Larger
Matchability criterion from Matchable Principle
nn
Sd
rd
rdN: Total number of cases, C1, C 2 , C 3 : Positive
constant
[Number of selected segments] rd →Make Smaller
Simplicityof Structure
Coveragefor Data
Matchability Focuses in covariance
Types of Relations Coincidence
The relation to which two events happen simultaneously Covariance
The relation that another attribute value will also change if one attribute value changes
Matchability: Estimates covariance
in selected data for categorical attributes.
A B C
ⅰ 80 10 10
ⅱ 10 80 10
ⅲ 10 10 80
How to find situations
Algorithms searches local maximums of Matchability Criterion.
Algorithm Overview for each subset of d of D Search Local maximums Reject saddle point end
Time complexity 2∝ D
Simple example
Input situation Mixture of cases on two
plains. Situation A: x + z = 1 Situation B: y +z = 1
Extracted situation Input Situations
MS 1= Input Situation A MS 2= Input Situation B
A New Situation MS 3 :
line x = y, x + z = 1
Generalization ability Multi-valued function φ:(x,y)→z
Even if the input situation A (x+z=1) lacks half of its parts, such that no data exists in the range y>0.5, our method outputs φMS1(0,1)=1.0.
Applications of Situation Decomposition (SD)
Multi-module Prediction System
Input Output
●Training cases
500 cases are sprayed on each plain in uniform distribution in the range x=[0.0, 1.0] and y=[0.0, 1.0].
●Test cases
11×11 cases are arranged to notches at a regular interval of 0.1 on each plane
Training cases and Test cases
q: sampling rate
1.E- 04
1.E- 03
1.E- 02
1.E- 010 20 40 60 80 100
Sampling rate q
Err
orPrediction Result
without Matchable Situation
with Matchable Situation
Autonomous Learning: AL Two step learning (Research in RWC)
Task learningExisting Knowledge
Acquiring environmental
knowledge
General factFor design
Acquiring solution for goal
goal
No reachingover the wall
Acquiringmovable paths
Generatingpath to the goal
Environment is given
Goal is given
Pre-task learning
DevelopmentToday’s Topic
Demonstration of Autonomous Learning
Door & Key task with CITTA
Start
Mobile Agent
Door
Telephone
Key
Goal
Agent acquire knowledge as situation
Door can open by the key.
Input/Output
Each Situation is used as Module
Position Action Object Belongings
Matchable Situation iMatchable Situation i
Matchable Situation iMatchable Situation i
Matchable Situation 1
Go by wallGo straight
Matchable Situation 2
Open door by telephone
Open door by Key
ExtractingMatchable Situation
Pre-task Learning
Combining Matchable Situation
Task Learning
…
Environment
Mobile Agent
Situation Decomposition in AL
SD in Pre-task learning:Situation decomposition handles input /output
vector of two time step for extracts Markov process.
Advantages by SD in Task learning: Adaptation by combining situations are
possible.Learning data can be reduced, because
learning space for each module is reduced.
Conclusions and Future works
Autonomous Learning
Conclusions
Pre-task learning
Self-organizingInformation Selection
Situation decomposition
Task learning
Cognitive Development
Situation Decomposition using Matchability Criterion
Matchable Principle
Matchability Criterion
Conclusions & Future workSituation decomposition Matchability is new model selection criterion maximizing matching
opportunity, which emphasize Coverage for data. In opposition ockham’s razor emphasize the Consistency for data. Decomposed situations by matchability criterion has powerful prediction ability. Situation decomposition method can be applied to pre-processing of data analysis, self-organization, pattern recognition and so on.
Future work
Situation decomposition: Needs theoretical research on Matchabilty criterion.
This intuitively delivered criterion affected unbalanced data.
Needs speed up for large-scale problem. Exponential time complexity for number of attribute is awful.
Advanced Self-organized Information Selection Situation decomposition method only selects set of attributes
and cases
Autonomous Learning: Relates with the knowledge of cognitive science.