SlideShare a Scribd company logo
1 of 12
German Terrazas [email_address] Dario Landa-Silva [email_address] Natalio Krasnogor [email_address] IV NICSO  May 12 – 14, 2010 Discovering Beneficial Cooperative Structures for the Automated Construction of Heuristics Extracted from: Information Génomique et Structurale – CNRS
Overview ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Hyper-heuristics ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
Hyper-heuristics Search methodologies choose low-level heuristics to solve hard computational problems Space of low-level heuristics Space of solutions selects  &  combines 120 fast & well performing Is it possible to automatically design the correct combination of low-level heuristics, the application of which results in good solutions for a given combinatorial optimisation problem ?  Combinatorial Optimisation Problem HOW TO  COMBINE ?
Q1b: How reliable are these combinations of low-level heuristics  ? Q1a: Given a set of high-level heuristics (which are combinations of low-level heuristics), is it possible to generate common combinations of low-level heuristics ? Q2: What is the performance of these  combinations when applied to the validation set  ? Q3a: Can pattern-based  heuristics be characterised by a template ? Q3b: What is the performance of the  template instances when applied to the test set ? Evaluation  and  filtering Randomly  created  heuristics Patterns  identification Pattern-based  heuristics  construction Pattern-based created  heuristics Template creation Pattern-based distilling Pattern-based  Heuristics  Generation Cross  Validation Template-based  Heuristics Distilling 1 2 3 P R O B L E M  Test dataset Validation dataset Training dataset TEMPLATE FOR  
Proof of Concept: STSP ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],2X 1CI 2X
10 different training tours    10 different pattern-based heuristics Q1b: How reliable are these combinations of low-level heuristics  ? 1 Information sharing Beneficial local search strategies Patterns  identification Pattern-based  heuristic construction Evaluation  and  filtering Best five heuristics kroA100_0.35612 EHHGTHHGHHTGTHHDDHDH kroA100_0.43440 ADGDADTDTHDDDCDD kroA100_0.45038 DHGHGACCCHCCACADC kroA100_0.46240 GHGHHGD kroA100_0.48562 GEGHGDD kroA100_1.46475 TGFCCGC kroA100_1.66957 AAHFFAFCFFFCGTG kroA100_2.34230 TGHGHHDHDHH kroA100_2.46724 CHCCECEHFGCFCF kroA100_2.55469 ATHGAGCDT Bottom worst  heuristics A C D E T F G H CHCCECEHFGCFCF  GEGHGDD  TGHGHHDHDHH  AAHFFAFCFFFCGTG  EHHGTHHGHHTGTHHDDHDH TGFCCGC  GHGHHGD  ATHGAGCDT  ADGDADTDTHDDDCDD  DHGHGACCCHCCACADC  Randomly  created  heuristics Q1a: Given a set of high-level heuristics, is it possible to generate common combinations of low-level heuristics ? 3OPT 2OPT 2X OROPT 1CI NI AI IO Applications of 300 randomly created heuristics Applications of  GDHGHHGDCDD Vs.
Q2: What is the performance of  these  combinations when applied  to the validation set  ? 2 For a given PBH and across the 2 nd  dataset (vkroaA100 i j  where i=75, j=0,…,9) 1) 300 COPIES OF PBH ( GEGHGDD ) 2) 300 RANDOMLY GENERATED HEURISTICS 3) MAX 30% SIMILARITY 4) 10 INDEP. EVALUATIONS OF 1 AND 2 300 randomly generated heuristics 300 copies of a given PBH
Template Q3a: Can pattern-based heuristics be characterised by a template ? 3 Common structures Building blocks C G F E G C D GA H C G FG C T TDGA D CH D D D D A CC DHH D F CD A D CTTTC C TD F CGG GD FAED HG EF CT TEE DGA D CH H D G D CHTE CC TFC D F CD ED D GGTFF C T F AHA G A D F H T G ED CT GH DGA HGH CHD E D GAF CC HC DCD EAH D TADD C DF F G G H D TGFF HG H CTD E GA EFE CH E D D D TE C A C D DC T D HFTE D FTCG CF TG GD T HG H CT CF D H GA A CH T DD ED CC DHC D T CD DDC D FD C EEH F AT GD GDAE H E G HHGD CT GFG DGA EEAH CHD H D A C E C A D C C T D TTDGF D H C E F A GD AA HGCT AE DGA TTA CHD C D HE C D CDCD CE D FEG Distilled Heuristics
Q3b: What is the performance of  the  template instances when  applied to the test dataset ? Across the 3 rd  dataset (gkroaA100 i j  where i=75, j=0,…,9) 1) 300 GRAMMAR GENERATED HEURISTICS 2) 300 RANDOMLY GENERATED HEURISTICS 3) MAX 30% SIMILARITY 4) 10 INDEP. EVALUATIONS OF 1 AND 2 300 randomly generated heuristics 300 grammar generated heuristics
Conclusions ,[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object],[object Object]
[object Object],[object Object],[object Object],[object Object],[object Object],German Terrazas [email_address] Dario Landa-Silva [email_address] Natalio Krasnogor [email_address] IV NICSO  May 12 – 14, 2010

More Related Content

Similar to Discovering Beneficial Cooperative Structures for the Automated Construction of Heuristics

Machine learning for_finance
Machine learning for_financeMachine learning for_finance
Machine learning for_financeStefan Duprey
 
A PSO-Based Subtractive Data Clustering Algorithm
A PSO-Based Subtractive Data Clustering AlgorithmA PSO-Based Subtractive Data Clustering Algorithm
A PSO-Based Subtractive Data Clustering AlgorithmIJORCS
 
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language ModelsDataScienceConferenc1
 
Towards the Design of Heuristics by Means of Self-Assembly
Towards the Design of Heuristics by Means of Self-AssemblyTowards the Design of Heuristics by Means of Self-Assembly
Towards the Design of Heuristics by Means of Self-AssemblyGerman Terrazas
 
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...Shubhashis Shil
 
Learning Content and Usage Factors Simultaneously
Learning Content and Usage Factors SimultaneouslyLearning Content and Usage Factors Simultaneously
Learning Content and Usage Factors SimultaneouslyArnab Bhadury
 
Efficient evaluation of flatness error from Coordinate Measurement Data using...
Efficient evaluation of flatness error from Coordinate Measurement Data using...Efficient evaluation of flatness error from Coordinate Measurement Data using...
Efficient evaluation of flatness error from Coordinate Measurement Data using...Ali Shahed
 
Research Away Day Jun 2009
Research Away Day Jun 2009Research Away Day Jun 2009
Research Away Day Jun 2009German Terrazas
 
G. Barcaroli, The use of machine learning in official statistics
G. Barcaroli, The use of machine learning in official statisticsG. Barcaroli, The use of machine learning in official statistics
G. Barcaroli, The use of machine learning in official statisticsIstituto nazionale di statistica
 
Icitam2019 2020 book_chapter
Icitam2019 2020 book_chapterIcitam2019 2020 book_chapter
Icitam2019 2020 book_chapterBan Bang
 
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...Vahid Taslimitehrani
 
An Adaptive Masker for the Differential Evolution Algorithm
An Adaptive Masker for the Differential Evolution AlgorithmAn Adaptive Masker for the Differential Evolution Algorithm
An Adaptive Masker for the Differential Evolution AlgorithmIOSR Journals
 
An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...Zac Darcy
 
An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...Zac Darcy
 
This is a heavily data-oriented
This is a heavily data-orientedThis is a heavily data-oriented
This is a heavily data-orientedbutest
 

Similar to Discovering Beneficial Cooperative Structures for the Automated Construction of Heuristics (20)

Cukoo srch
Cukoo srchCukoo srch
Cukoo srch
 
Cukoo srch
Cukoo srchCukoo srch
Cukoo srch
 
Machine learning for_finance
Machine learning for_financeMachine learning for_finance
Machine learning for_finance
 
ga-2.ppt
ga-2.pptga-2.ppt
ga-2.ppt
 
A PSO-Based Subtractive Data Clustering Algorithm
A PSO-Based Subtractive Data Clustering AlgorithmA PSO-Based Subtractive Data Clustering Algorithm
A PSO-Based Subtractive Data Clustering Algorithm
 
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models
[DSC Europe 23] Dmitry Ustalov - Design and Evaluation of Large Language Models
 
Kdd by Mr.Sameer Kumar Das
Kdd by Mr.Sameer Kumar DasKdd by Mr.Sameer Kumar Das
Kdd by Mr.Sameer Kumar Das
 
Towards the Design of Heuristics by Means of Self-Assembly
Towards the Design of Heuristics by Means of Self-AssemblyTowards the Design of Heuristics by Means of Self-Assembly
Towards the Design of Heuristics by Means of Self-Assembly
 
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...
Solving Multidimensional Multiple Choice Knapsack Problem By Genetic Algorith...
 
Learning Content and Usage Factors Simultaneously
Learning Content and Usage Factors SimultaneouslyLearning Content and Usage Factors Simultaneously
Learning Content and Usage Factors Simultaneously
 
P1121133727
P1121133727P1121133727
P1121133727
 
Efficient evaluation of flatness error from Coordinate Measurement Data using...
Efficient evaluation of flatness error from Coordinate Measurement Data using...Efficient evaluation of flatness error from Coordinate Measurement Data using...
Efficient evaluation of flatness error from Coordinate Measurement Data using...
 
Research Away Day Jun 2009
Research Away Day Jun 2009Research Away Day Jun 2009
Research Away Day Jun 2009
 
G. Barcaroli, The use of machine learning in official statistics
G. Barcaroli, The use of machine learning in official statisticsG. Barcaroli, The use of machine learning in official statistics
G. Barcaroli, The use of machine learning in official statistics
 
Icitam2019 2020 book_chapter
Icitam2019 2020 book_chapterIcitam2019 2020 book_chapter
Icitam2019 2020 book_chapter
 
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...
A new CPXR Based Logistic Regression Method and Clinical Prognostic Modeling ...
 
An Adaptive Masker for the Differential Evolution Algorithm
An Adaptive Masker for the Differential Evolution AlgorithmAn Adaptive Masker for the Differential Evolution Algorithm
An Adaptive Masker for the Differential Evolution Algorithm
 
An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...
 
An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...An Improved Iterative Method for Solving General System of Equations via Gene...
An Improved Iterative Method for Solving General System of Equations via Gene...
 
This is a heavily data-oriented
This is a heavily data-orientedThis is a heavily data-oriented
This is a heavily data-oriented
 

Recently uploaded

2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptxMaritesTamaniVerdade
 
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdfUnit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdfDr Vijay Vishwakarma
 
FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024Elizabeth Walsh
 
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...Pooja Bhuva
 
Plant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptxPlant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptxUmeshTimilsina1
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfagholdier
 
How to setup Pycharm environment for Odoo 17.pptx
How to setup Pycharm environment for Odoo 17.pptxHow to setup Pycharm environment for Odoo 17.pptx
How to setup Pycharm environment for Odoo 17.pptxCeline George
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfSherif Taha
 
Sociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning ExhibitSociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning Exhibitjbellavia9
 
REMIFENTANIL: An Ultra short acting opioid.pptx
REMIFENTANIL: An Ultra short acting opioid.pptxREMIFENTANIL: An Ultra short acting opioid.pptx
REMIFENTANIL: An Ultra short acting opioid.pptxDr. Ravikiran H M Gowda
 
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxBasic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxDenish Jangid
 
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptx
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptxHMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptx
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptxmarlenawright1
 
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...Amil baba
 
Wellbeing inclusion and digital dystopias.pptx
Wellbeing inclusion and digital dystopias.pptxWellbeing inclusion and digital dystopias.pptx
Wellbeing inclusion and digital dystopias.pptxJisc
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...ZurliaSoop
 
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...Pooja Bhuva
 
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptxOn_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptxPooja Bhuva
 
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptxCOMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptxannathomasp01
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.pptRamjanShidvankar
 
Single or Multiple melodic lines structure
Single or Multiple melodic lines structureSingle or Multiple melodic lines structure
Single or Multiple melodic lines structuredhanjurrannsibayan2
 

Recently uploaded (20)

2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
2024-NATIONAL-LEARNING-CAMP-AND-OTHER.pptx
 
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdfUnit 3 Emotional Intelligence and Spiritual Intelligence.pdf
Unit 3 Emotional Intelligence and Spiritual Intelligence.pdf
 
FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024FSB Advising Checklist - Orientation 2024
FSB Advising Checklist - Orientation 2024
 
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...
Beyond_Borders_Understanding_Anime_and_Manga_Fandom_A_Comprehensive_Audience_...
 
Plant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptxPlant propagation: Sexual and Asexual propapagation.pptx
Plant propagation: Sexual and Asexual propapagation.pptx
 
Holdier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdfHoldier Curriculum Vitae (April 2024).pdf
Holdier Curriculum Vitae (April 2024).pdf
 
How to setup Pycharm environment for Odoo 17.pptx
How to setup Pycharm environment for Odoo 17.pptxHow to setup Pycharm environment for Odoo 17.pptx
How to setup Pycharm environment for Odoo 17.pptx
 
Food safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdfFood safety_Challenges food safety laboratories_.pdf
Food safety_Challenges food safety laboratories_.pdf
 
Sociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning ExhibitSociology 101 Demonstration of Learning Exhibit
Sociology 101 Demonstration of Learning Exhibit
 
REMIFENTANIL: An Ultra short acting opioid.pptx
REMIFENTANIL: An Ultra short acting opioid.pptxREMIFENTANIL: An Ultra short acting opioid.pptx
REMIFENTANIL: An Ultra short acting opioid.pptx
 
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptxBasic Civil Engineering first year Notes- Chapter 4 Building.pptx
Basic Civil Engineering first year Notes- Chapter 4 Building.pptx
 
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptx
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptxHMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptx
HMCS Vancouver Pre-Deployment Brief - May 2024 (Web Version).pptx
 
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...
NO1 Top Black Magic Specialist In Lahore Black magic In Pakistan Kala Ilam Ex...
 
Wellbeing inclusion and digital dystopias.pptx
Wellbeing inclusion and digital dystopias.pptxWellbeing inclusion and digital dystopias.pptx
Wellbeing inclusion and digital dystopias.pptx
 
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
Jual Obat Aborsi Hongkong ( Asli No.1 ) 085657271886 Obat Penggugur Kandungan...
 
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...
Sensory_Experience_and_Emotional_Resonance_in_Gabriel_Okaras_The_Piano_and_Th...
 
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptxOn_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
On_Translating_a_Tamil_Poem_by_A_K_Ramanujan.pptx
 
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptxCOMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
COMMUNICATING NEGATIVE NEWS - APPROACHES .pptx
 
Application orientated numerical on hev.ppt
Application orientated numerical on hev.pptApplication orientated numerical on hev.ppt
Application orientated numerical on hev.ppt
 
Single or Multiple melodic lines structure
Single or Multiple melodic lines structureSingle or Multiple melodic lines structure
Single or Multiple melodic lines structure
 

Discovering Beneficial Cooperative Structures for the Automated Construction of Heuristics

  • 1. German Terrazas [email_address] Dario Landa-Silva [email_address] Natalio Krasnogor [email_address] IV NICSO May 12 – 14, 2010 Discovering Beneficial Cooperative Structures for the Automated Construction of Heuristics Extracted from: Information Génomique et Structurale – CNRS
  • 2.
  • 3.
  • 4. Hyper-heuristics Search methodologies choose low-level heuristics to solve hard computational problems Space of low-level heuristics Space of solutions selects & combines 120 fast & well performing Is it possible to automatically design the correct combination of low-level heuristics, the application of which results in good solutions for a given combinatorial optimisation problem ? Combinatorial Optimisation Problem HOW TO COMBINE ?
  • 5. Q1b: How reliable are these combinations of low-level heuristics ? Q1a: Given a set of high-level heuristics (which are combinations of low-level heuristics), is it possible to generate common combinations of low-level heuristics ? Q2: What is the performance of these combinations when applied to the validation set ? Q3a: Can pattern-based heuristics be characterised by a template ? Q3b: What is the performance of the template instances when applied to the test set ? Evaluation and filtering Randomly created heuristics Patterns identification Pattern-based heuristics construction Pattern-based created heuristics Template creation Pattern-based distilling Pattern-based Heuristics Generation Cross Validation Template-based Heuristics Distilling 1 2 3 P R O B L E M  Test dataset Validation dataset Training dataset TEMPLATE FOR 
  • 6.
  • 7. 10 different training tours  10 different pattern-based heuristics Q1b: How reliable are these combinations of low-level heuristics ? 1 Information sharing Beneficial local search strategies Patterns identification Pattern-based heuristic construction Evaluation and filtering Best five heuristics kroA100_0.35612 EHHGTHHGHHTGTHHDDHDH kroA100_0.43440 ADGDADTDTHDDDCDD kroA100_0.45038 DHGHGACCCHCCACADC kroA100_0.46240 GHGHHGD kroA100_0.48562 GEGHGDD kroA100_1.46475 TGFCCGC kroA100_1.66957 AAHFFAFCFFFCGTG kroA100_2.34230 TGHGHHDHDHH kroA100_2.46724 CHCCECEHFGCFCF kroA100_2.55469 ATHGAGCDT Bottom worst heuristics A C D E T F G H CHCCECEHFGCFCF GEGHGDD TGHGHHDHDHH AAHFFAFCFFFCGTG EHHGTHHGHHTGTHHDDHDH TGFCCGC GHGHHGD ATHGAGCDT ADGDADTDTHDDDCDD DHGHGACCCHCCACADC Randomly created heuristics Q1a: Given a set of high-level heuristics, is it possible to generate common combinations of low-level heuristics ? 3OPT 2OPT 2X OROPT 1CI NI AI IO Applications of 300 randomly created heuristics Applications of GDHGHHGDCDD Vs.
  • 8. Q2: What is the performance of these combinations when applied to the validation set ? 2 For a given PBH and across the 2 nd dataset (vkroaA100 i j where i=75, j=0,…,9) 1) 300 COPIES OF PBH ( GEGHGDD ) 2) 300 RANDOMLY GENERATED HEURISTICS 3) MAX 30% SIMILARITY 4) 10 INDEP. EVALUATIONS OF 1 AND 2 300 randomly generated heuristics 300 copies of a given PBH
  • 9. Template Q3a: Can pattern-based heuristics be characterised by a template ? 3 Common structures Building blocks C G F E G C D GA H C G FG C T TDGA D CH D D D D A CC DHH D F CD A D CTTTC C TD F CGG GD FAED HG EF CT TEE DGA D CH H D G D CHTE CC TFC D F CD ED D GGTFF C T F AHA G A D F H T G ED CT GH DGA HGH CHD E D GAF CC HC DCD EAH D TADD C DF F G G H D TGFF HG H CTD E GA EFE CH E D D D TE C A C D DC T D HFTE D FTCG CF TG GD T HG H CT CF D H GA A CH T DD ED CC DHC D T CD DDC D FD C EEH F AT GD GDAE H E G HHGD CT GFG DGA EEAH CHD H D A C E C A D C C T D TTDGF D H C E F A GD AA HGCT AE DGA TTA CHD C D HE C D CDCD CE D FEG Distilled Heuristics
  • 10. Q3b: What is the performance of the template instances when applied to the test dataset ? Across the 3 rd dataset (gkroaA100 i j where i=75, j=0,…,9) 1) 300 GRAMMAR GENERATED HEURISTICS 2) 300 RANDOMLY GENERATED HEURISTICS 3) MAX 30% SIMILARITY 4) 10 INDEP. EVALUATIONS OF 1 AND 2 300 randomly generated heuristics 300 grammar generated heuristics
  • 11.
  • 12.