XML-based Genetic Programming Framework: Design Philosophy, Implementation and Applications

DOSHISHA UNIVERSITY

April 19, 2023 1

XML-based Genetic Programming Framework:Design Philosophy, Implementation

and Applications

DOSHISHA UNIVERSITY

April 19, 2023 2

Outline

1.Introduction2. Objective3. Proposed approach4. Verification results5. Applications6. Conclusion

DOSHISHA UNIVERSITY

April 19, 2023 3

1. Introduction: the Problem

The NeedsThe Needs

A) Promptly developed software models of the evolved artifacts B) Fast running offline (phylogenetic) learning via simulated evolution (e.g. GP)


DOSHISHA UNIVERSITY

April 19, 2023 4

The NeedsThe Needs



The RealityThe Reality

A) Slow development time of evolutionary systems (specific semantics)B) Notoriously poor performance of GP (populations, generations, independent runs)


Discrepancy, Gap Discrepancy, Gap

1. Introduction: the Problem

DOSHISHA UNIVERSITY

April 19, 2023 5

Discrepancy, Gap

The NeedsThe Needs

A) Promptly developed software agents B) Fast running offline (phylogenetic) learning via simulated evolution (e.g. GP)

A) Promptly developed software agents B) Fast running offline (phylogenetic) learning via simulated evolution (e.g. GP)




A) Quicker development time GP B) Better performance characteristics of GP

A) Quicker development time GP B) Better performance characteristics of GP


2. The Objective

DOSHISHA UNIVERSITY

April 19, 2023 6

3. The Approach

Quicker development time of GP ?Quicker development time of GP ?

History of “Reuse of Software Blocks” in Software Engineering:

• loops,• procedures, functions (incl. recursions),• modules (units),• objects,• component objects

History of “Reuse of Software Blocks” in Software Engineering:

• loops,• procedures, functions (incl. recursions),• modules (units),• objects,• component objects

Component objects (CO): • appears to be an object of the IDE which incorporates them,• binary standard (language-independent)

Component objects (CO): • appears to be an object of the IDE which incorporates them,• binary standard (language-independent)

DOSHISHA UNIVERSITY

April 19, 2023 7

3. The Approach

Focusing on representation of genetic programs:Focusing on representation of genetic programs:

A) Standard DOM-parsing tree and XML text.

A) Standard DOM-parsing tree and XML text.

B) CO: DOM-parser with built-in API for dealing with genetic programs.

B) CO: DOM-parser with built-in API for dealing with genetic programs.

DOSHISHA UNIVERSITY

April 19, 2023 8

3. The Approach

AdvantagesAdvantages

A) Significant reduction of the time consumption of software engineering of GP using build-in API for creating and manipulating genetic programs.

A) Significant reduction of the time consumption of software engineering of GP using build-in API for creating and manipulating genetic programs.

DOSHISHA UNIVERSITY

April 19, 2023 9

3. The Approach

Issue: How to represent the allowed syntax (i.e. to reduce the search space) of GP?

• In the program source of GP-system (modifications by expert, recompilation, etc…) ?• As an external text with well-known format?

Employing XML facilitates the second choice.

Issue: How to represent the allowed syntax (i.e. to reduce the search space) of GP?

• In the program source of GP-system (modifications by expert, recompilation, etc…) ?• As an external text with well-known format?

Employing XML facilitates the second choice.

DOSHISHA UNIVERSITY

April 19, 2023 10

3. The Approach

B) Increase of efficiency of execution of XGP:

Reducing the computational effort as a result of

generic support for the idea of pruning the solution

space via strongly typed GP.

How:

XML-schema as a standard, generic way to represent

the syntax of XGP.

B) Increase of efficiency of execution of XGP:

Reducing the computational effort as a result of

generic support for the idea of pruning the solution

space via strongly typed GP.

How:

XML-schema as a standard, generic way to represent

the syntax of XGP.


DOSHISHA UNIVERSITY

April 19, 2023 11

3. The Approach

• Relationship between tree nodes in XGP,• Data types associated with tree nodes

• Relationship between tree nodes in XGP,• Data types associated with tree nodes

<xs:simpleType name="VAR_TSpeed"><xs:restriction base="xs:string"> <xs:enumeration value=“Speed" /> </xs:restriction></xs:simpleType><xs:simpleType name="OPER_TSpeed"><xs:restriction base="xs:string"> <xs:enumeration value="GE" /> <xs:enumeration value="LE" /> </xs:restriction></xs:simpleType><xs:simpleType name="CONST_TSpeed"><xs:restriction base="xs:integer"> <xs:minInclusive value="0" /> <xs:maxInclusive value=“22" /> </xs:restriction></xs:simpleType>

Fragment of XML Schema

DOSHISHA UNIVERSITY

April 19, 2023 12

3. The Approach

B) Increase of efficiency of execution of XGP - parallelism: • Improving the computational performance: XML representation of both the schema and the genetic programs is a feasible format for migration of agents in parallel, distributed computer architectures.

B) Increase of efficiency of execution of XGP - parallelism: • Improving the computational performance: XML representation of both the schema and the genetic programs is a feasible format for migration of agents in parallel, distributed computer architectures.


In-memory tree structures of GP cannot be transferred between computing units in parallel architectures.

In-memory tree structures of GP cannot be transferred between computing units in parallel architectures.

DOSHISHA UNIVERSITY

April 19, 2023 13

3. The Approach

Memory Structure (DOM)Text (XML)

StraightforwardMapping

DOSHISHA UNIVERSITY

April 19, 2023 14

3. The Approach

GP Manager (selection, crossover, and mutation)

Domain Independent (only XML Schema need to be

updated)

Simulation Boards (evaluation)

Domain-specific

Structure of XGP-frameworkStructure of XGP-framework

Implications:• Reuse of GP Manager across the applications,• Parallel Simulation Boards

DOSHISHA UNIVERSITY

April 19, 2023 15

Parallel Implementation via Boss-Workers ModelExample – Evolution of Behavior of Agents in MAS

GP Manager(selection, crossover,

and mutation)

GP Manager(selection, crossover,

and mutation)Simulation Boards

(evaluation)

Simulation Boards(evaluation)

Genetic program (XML)

Fitness

3. The Approach

DOSHISHA UNIVERSITY

April 19, 2023 16

4. Verification Results

• Development time for the initial prototype of XGP (from scratch): several [person*days]

• Development time for the initial prototype of XGP (from scratch): several [person*days]

DOSHISHA UNIVERSITY

April 19, 2023 17


• Porting time (employing XGP for already developed simulation board): less than one hour

• Porting time (employing XGP for already developed simulation board): less than one hour

XML SchemaFile

DOSHISHA UNIVERSITY

April 19, 2023 18


• Computational Effort of XGP: Reducing the search Space (XML Schema)

• Computational Effort of XGP: Reducing the search Space (XML Schema)

0.0

0.2

0.4

0.6

0.8

1.0

0 8000 16000 24000 32000 40000Indiv iduals ev aluated

p(t)

STGPLPLPA

Probability of Success for Evolution of XGP with (STGP) and without (LP, LPA) strong types

Probability of Success for Evolution of XGP with (STGP) and without (LP, LPA) strong types

DOSHISHA UNIVERSITY

April 19, 2023 19

GP ManagerDomain Neutral

MAS Simulation Board

Domain Specific

5. Applications

Evolution of Agents Behavior in MASEvolution of Agents Behavior in MAS

DOSHISHA UNIVERSITY

April 19, 2023 20

XML representation

of GP

5. Applications

Evolution of Agents Behavior in MASEvolution of Agents Behavior in MAS

DOSHISHA UNIVERSITY

April 19, 2023 21


Simulation BoardDomain Specific

5. Applications

DOM representation

of GP

Evolution of Locomotion of SnakebotEvolution of Locomotion of Snakebot

DOSHISHA UNIVERSITY

April 19, 2023 22



5. Applications

XML representation

of GP

Evolution of Neural NetworksEvolution of Neural Networks

DOSHISHA UNIVERSITY

April 19, 2023 23

Car(1/24 Scale Model)

Remote Control(agent’s actions)

Camera(perceptionsof the agent)

PC(driving agent)

ControlLoop, 100ms

5. Applications

Evolution of Driving AgentEvolution of Driving Agent

DOSHISHA UNIVERSITY

April 19, 2023 24



5. Applications

DOM representation

of GP

Evolution of Driving AgentEvolution of Driving Agent

DOSHISHA UNIVERSITY

April 19, 2023 25



5. Applications

DOM representation

of GP

Interactive Evolution of Postures of Aibo RobotInteractive Evolution of Postures of Aibo Robot

DOSHISHA UNIVERSITY

April 19, 2023 26



5. Applications

DOM representation

of GP

Interactive Evolution of Room ColorsInteractive Evolution of Room Colors

DOSHISHA UNIVERSITY

April 19, 2023 27



5. Applications

Evolution of Human-Relation NetworksEvolution of Human-Relation Networks

DOSHISHA UNIVERSITY

April 19, 2023 28

6. Conclusion

A)Reduced Development Time• Managing genetic program via standard DOM

parsers with built-in API

Proposed DOM/XML-Based Portable Genetic Representation in XGP

B) Easy Porting to New Applications• Reusing the very General, Domain-Independent

GP Manager,• Modifying the XML-schema only.

DOSHISHA UNIVERSITY

April 19, 2023 29

6. Conclusion

Proposed DOM/XML-Based Portable Genetic Representation in XGP

C) Improved Execution Time of XGP• Reducing Computational Effort: Limiting solution

space using strongly typed GP and offering generic support via XML schema,

• Improving Computational Performance: Generic support of distributed (web-compliant) implementation of GP.

Drawbacks?• Fitness evaluation – parsing of XML/DOM tree and navigating among the nodes…

XML-based Genetic Programming Framework: Design Philosophy, Implementation and Applications

Documents

evolution of xgp

typed gp

execution of xgp

syntax of xgp

search space of gp

reuse of gp manager

program source of gp

xml representation of