SlideShare una empresa de Scribd logo
1 de 28
Bandit­based EDA for noisy
optimization.


    P. Rolet, O. Teytaud
        Birmingham, 2009



Tao, Inria Saclay Ile-De-France, LRI (Université Paris Sud, France),
UMR CNRS 8623, I&A team, Digiteo, Pascal Network of Excellence.
Outline




   Introduction
   Lower bound
   An intuitive solution: bandits
   Rigorous upper-bound and rigorous solution




  Rolet and Teytaud           TRSH 09 is great   2
Introduction




   Noisy optimization:
   - the fitness function returns a noisy answer;
   - this noisy answer is independent;
   - goal: finding best expected value.




  Rolet and Teytaud             TRSH 09 is great    3
Main argument of this work

 Usual algorithms don't work in noisy optimization.

 The computational
 power is like this ==>

                     <== and the result is like that
                     because algos are not consistent.

 ==> there is much to win, algorithms can be
 greatly improved.

 Rolet and Teytaud                         TRSH 09 is great   4
Previous works




  Should we average or not ?
  Or should we increase lambda ?
   ==> various answers in the literature.




 Rolet and Teytaud           TRSH 09 is great   5
Jebalia, Auger, PPSN 2008




   .
   .

 ==> scale-invariant 1+1-ES
       converges linearly



 Rolet and Teytaud            TRSH 09 is great   6
Jebalia, Auger, PPSN 2008



   .
   .

 In the present work:
 - no lower-bound on fnoise(x)/f(x)
 - “real” algorithm (not scale-invariant)
 ==> but slower rate (yet tight)
 Rolet and Teytaud             TRSH 09 is great   7
Introduction




   Introduction
   Lower bound
   An intuitive solution: bandits
   Rigorous upper-bound and rigorous solution




  Rolet and Teytaud           TRSH 09 is great   8
Lower bound: boring framework




                                       Optimization
                                        algorithm




 Rolet and Teytaud              TRSH 09 is great      9
Lower bound: boring framework




                                Noisy (binary) measurement


                     It's a lower bound. If it holds in the binary
                     case, it holds in the non-binary case either.
 Rolet and Teytaud                             TRSH 09 is great      10
Lower bound: boring framework




                                 Final loss

 Rolet and Teytaud              TRSH 09 is great   11
How to prove a lower bound ? Simple case...


    Consider we are in dimension 2. Consider that you
    solve the problem with precision .

    Consider a regular simplex of possible optima:

              d(ti,tj)=




    (here, for simplicity: deterministic algorithm)


 Rolet and Teytaud                     TRSH 09 is great   12
How to prove a lower bound ? Simple case...

    Visit point x

    f(x,ti) = f(x,tj) ±


    so with proba 1-,
    f   (x,ti) = f  (x,tj)
     noise           noise


    ==> with proba 1-,
        an iteration with optimum in ti
      = an iteration with optimum in tj


 Rolet and Teytaud                        TRSH 09 is great   13
How to prove a lower bound ? Simple case...

With proba 1-,
    an iteration with optimum in ti
  = an iteration with optimum in tj

With proba 1-N,
    a run with optimum in ti
 = a run with optimum in tj




==> real case = similar to this one.
 Rolet and Teytaud                    TRSH 09 is great   14
Lower bound




 Rolet and Teytaud   TRSH 09 is great   15
Introduction




   Introduction
   Lower bound
   An intuitive solution: bandits
   Rigorous upper-bound and rigorous solution




  Rolet and Teytaud           TRSH 09 is great   16
Idea of bandits


 I have N arms. I have t time steps.
 Pulling an arm yields a reward in [0,1].
 Each arm has a stationary (i.i.d) reward.
 At each time step I can pull an arm.

 How will I find good arms ?




  Rolet and Teytaud             TRSH 09 is great   17
Idea of bandits



 The goal of Bernstein races:
 - guessing which arms are the most rewarding;
 - whilst saving up time.




  Rolet and Teytaud           TRSH 09 is great   18
Bernstein race: general idea.

 While (I want to go on)
 {
   I pull once each non discarded arm.
   I compute a lower and upper bound for all
        non-discarded arms (Bernstein bound).
   I discard arms which are excluded by the
        bounds.
 }


  Rolet and Teytaud             TRSH 09 is great   19
Already used for noisy optimization




   Idea:
     - evolutionary algorithm (CMA)
     - replacing selection by Bernstein race:
            keep racing until  points are selected.

   Trouble:
     - sometimes very expensive iterations
     - tricks are added in the algorithm, not stable


  Rolet and Teytaud                     TRSH 09 is great   20
Our version
  Idea:
    - evolutionary algorithm;
    - derandomized mutation ensuring that there are
        at least two points with “sufficiently different”
        values.
    - replacing selection by Bernstein race:
           keep racing until  points are significantly
           better than 'other points. (=1, '=1)

  Tricky part: derandomized mutation (one more step in
      the algorithm; less simple);

  Otherwise no trouble, proved, we don't have
      to add tricks.
 Rolet and Teytaud                       TRSH 09 is great   21
Introduction




  Introduction
  Lower bound
  An intuitive solution: bandits
  Rigorous upper-bound and rigorous
  solution



  Rolet and Teytaud       TRSH 09 is great   22
Bernstein race for comparing two among three
arms with confidence 1-'




 Rolet and Teytaud            TRSH 09 is great   23
Bernstein race for comparing two among three
arms with confidence 1-'



                               Bernstein principle
                               (low variance)




 Rolet and Teytaud            TRSH 09 is great       24
Bernstein race for comparing 2 among 3 arms with
confidence 1-'




                                    THE important part:
                                    points are
                                    signif. different!




  Rolet and Teytaud               TRSH 09 is great    25
The complete algorithm
 The algorithm iteratively improves the set of
 possible optima. Iteration n as follows:

 1) Generate 3 points (equally spaced on a line).

 2) Apply the Bernstein race until one of the arms (the
    good arm) is statistically better than at least one
   other arm (the bad arm) for some ' = O(1/n2).

 3) Remove the part of the domain which is farther
    from the good arm than from the bad arm.

 Sum of the ' =  ==> proof with confidence 1-.

 Rolet and Teytaud                    TRSH 09 is great    26
The complete algorithm




 Rolet and Teytaud       TRSH 09 is great   27
Conclusion
Bandits = good tool for noisy optimization.

 But taking care of how to generate points is necessary
(otherwise one might have points with very similar
fitness values which is dangerous!).

Further work:
* Generalizing the approach to more simple algorithms.

* Lower bound = upper bound up to constant factors
                        depending on the dimension

* Proof of slower rate with variance not decreasing
   to zero at the optimum.

  Rolet and Teytaud                    TRSH 09 is great   28

Más contenido relacionado

La actualidad más candente

Pattern Matching Part Three: Hamming Distance
Pattern Matching Part Three: Hamming DistancePattern Matching Part Three: Hamming Distance
Pattern Matching Part Three: Hamming DistanceBenjamin Sach
 
Pattern Matching Part Two: k-mismatches
Pattern Matching Part Two: k-mismatchesPattern Matching Part Two: k-mismatches
Pattern Matching Part Two: k-mismatchesBenjamin Sach
 
Asymptotic Notations
Asymptotic NotationsAsymptotic Notations
Asymptotic NotationsNagendraK18
 
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties Dr.SHANTHI K.G
 
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...Amrinder Arora
 
Introduction to Digital Signal Processing (DSP) - Course Notes
Introduction to Digital Signal Processing (DSP) - Course NotesIntroduction to Digital Signal Processing (DSP) - Course Notes
Introduction to Digital Signal Processing (DSP) - Course NotesAhmed Gad
 
Pattern Matching Part One: Suffix Trees
Pattern Matching Part One: Suffix TreesPattern Matching Part One: Suffix Trees
Pattern Matching Part One: Suffix TreesBenjamin Sach
 
Computability - Tractable, Intractable and Non-computable Function
Computability - Tractable, Intractable and Non-computable FunctionComputability - Tractable, Intractable and Non-computable Function
Computability - Tractable, Intractable and Non-computable FunctionReggie Niccolo Santos
 
Fourier-transform analysis of a unilateral fin line and its derivatives
Fourier-transform analysis of a unilateral fin line and its derivativesFourier-transform analysis of a unilateral fin line and its derivatives
Fourier-transform analysis of a unilateral fin line and its derivativesYong Heui Cho
 
Fourier analysis
Fourier analysisFourier analysis
Fourier analysisAyzaJabeen
 
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS Dr.SHANTHI K.G
 
Turing Machine
Turing MachineTuring Machine
Turing MachineRajendran
 

La actualidad más candente (18)

smtlectures.1
smtlectures.1smtlectures.1
smtlectures.1
 
Pattern Matching Part Three: Hamming Distance
Pattern Matching Part Three: Hamming DistancePattern Matching Part Three: Hamming Distance
Pattern Matching Part Three: Hamming Distance
 
Chapter 2 fourier transform
Chapter 2 fourier transformChapter 2 fourier transform
Chapter 2 fourier transform
 
Pattern Matching Part Two: k-mismatches
Pattern Matching Part Two: k-mismatchesPattern Matching Part Two: k-mismatches
Pattern Matching Part Two: k-mismatches
 
Asymptotic Notations
Asymptotic NotationsAsymptotic Notations
Asymptotic Notations
 
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties
Fourier Transform ,LAPLACE TRANSFORM,ROC and its Properties
 
Turing machine
Turing machineTuring machine
Turing machine
 
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...
Bron Kerbosch Algorithm - Presentation by Jun Zhai, Tianhang Qiang and Yizhen...
 
Introduction to Digital Signal Processing (DSP) - Course Notes
Introduction to Digital Signal Processing (DSP) - Course NotesIntroduction to Digital Signal Processing (DSP) - Course Notes
Introduction to Digital Signal Processing (DSP) - Course Notes
 
smtlecture.4
smtlecture.4smtlecture.4
smtlecture.4
 
Pattern Matching Part One: Suffix Trees
Pattern Matching Part One: Suffix TreesPattern Matching Part One: Suffix Trees
Pattern Matching Part One: Suffix Trees
 
Computability - Tractable, Intractable and Non-computable Function
Computability - Tractable, Intractable and Non-computable FunctionComputability - Tractable, Intractable and Non-computable Function
Computability - Tractable, Intractable and Non-computable Function
 
Fourier-transform analysis of a unilateral fin line and its derivatives
Fourier-transform analysis of a unilateral fin line and its derivativesFourier-transform analysis of a unilateral fin line and its derivatives
Fourier-transform analysis of a unilateral fin line and its derivatives
 
Fourier analysis
Fourier analysisFourier analysis
Fourier analysis
 
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS
REVISION- UNIT 2 -ANALYSIS OF CONTINUOUS TIME SIGNALS
 
Turing Machine
Turing MachineTuring Machine
Turing Machine
 
Turing machine
Turing machineTuring machine
Turing machine
 
Signal &amp; system
Signal &amp; systemSignal &amp; system
Signal &amp; system
 

Destacado

Ilab Metis: we optimize power systems and we are not afraid of direct policy ...
Ilab Metis: we optimize power systems and we are not afraid of direct policy ...Ilab Metis: we optimize power systems and we are not afraid of direct policy ...
Ilab Metis: we optimize power systems and we are not afraid of direct policy ...Olivier Teytaud
 
Artificial intelligence and blind Go
Artificial intelligence and blind GoArtificial intelligence and blind Go
Artificial intelligence and blind GoOlivier Teytaud
 
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...Olivier Teytaud
 
Games with partial information
Games with partial informationGames with partial information
Games with partial informationOlivier Teytaud
 
Provocative statements around energy
Provocative statements around energyProvocative statements around energy
Provocative statements around energyOlivier Teytaud
 
Tools for Discrete Time Control; Application to Power Systems
Tools for Discrete Time Control; Application to Power SystemsTools for Discrete Time Control; Application to Power Systems
Tools for Discrete Time Control; Application to Power SystemsOlivier Teytaud
 
Choosing between several options in uncertain environments
Choosing between several options in uncertain environmentsChoosing between several options in uncertain environments
Choosing between several options in uncertain environmentsOlivier Teytaud
 
Energy Management (production side)
Energy Management (production side)Energy Management (production side)
Energy Management (production side)Olivier Teytaud
 
The game of Go and energy; two nice computational intelligence problems (with...
The game of Go and energy; two nice computational intelligence problems (with...The game of Go and energy; two nice computational intelligence problems (with...
The game of Go and energy; two nice computational intelligence problems (with...Olivier Teytaud
 
Optimization of power systems - old and new tools
Optimization of power systems - old and new toolsOptimization of power systems - old and new tools
Optimization of power systems - old and new toolsOlivier Teytaud
 
Artificial Intelligence and Optimization with Parallelism
Artificial Intelligence and Optimization with ParallelismArtificial Intelligence and Optimization with Parallelism
Artificial Intelligence and Optimization with ParallelismOlivier Teytaud
 

Destacado (17)

Hydroelectricity
HydroelectricityHydroelectricity
Hydroelectricity
 
Openoffice and Linux
Openoffice and LinuxOpenoffice and Linux
Openoffice and Linux
 
Ilab Metis: we optimize power systems and we are not afraid of direct policy ...
Ilab Metis: we optimize power systems and we are not afraid of direct policy ...Ilab Metis: we optimize power systems and we are not afraid of direct policy ...
Ilab Metis: we optimize power systems and we are not afraid of direct policy ...
 
Artificial intelligence and blind Go
Artificial intelligence and blind GoArtificial intelligence and blind Go
Artificial intelligence and blind Go
 
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...
Tools for artificial intelligence: EXP3, Zermelo algorithm, Alpha-Beta, and s...
 
Games with partial information
Games with partial informationGames with partial information
Games with partial information
 
Tutorialmcts
TutorialmctsTutorialmcts
Tutorialmcts
 
Provocative statements around energy
Provocative statements around energyProvocative statements around energy
Provocative statements around energy
 
Tools for Discrete Time Control; Application to Power Systems
Tools for Discrete Time Control; Application to Power SystemsTools for Discrete Time Control; Application to Power Systems
Tools for Discrete Time Control; Application to Power Systems
 
Choosing between several options in uncertain environments
Choosing between several options in uncertain environmentsChoosing between several options in uncertain environments
Choosing between several options in uncertain environments
 
Energy Management (production side)
Energy Management (production side)Energy Management (production side)
Energy Management (production side)
 
The game of Go and energy; two nice computational intelligence problems (with...
The game of Go and energy; two nice computational intelligence problems (with...The game of Go and energy; two nice computational intelligence problems (with...
The game of Go and energy; two nice computational intelligence problems (with...
 
Theory of games
Theory of gamesTheory of games
Theory of games
 
Optimization of power systems - old and new tools
Optimization of power systems - old and new toolsOptimization of power systems - old and new tools
Optimization of power systems - old and new tools
 
Grenoble
GrenobleGrenoble
Grenoble
 
Labex2012g
Labex2012gLabex2012g
Labex2012g
 
Artificial Intelligence and Optimization with Parallelism
Artificial Intelligence and Optimization with ParallelismArtificial Intelligence and Optimization with Parallelism
Artificial Intelligence and Optimization with Parallelism
 

Similar a Noisy Optimization combining Bandits and Evolutionary Algorithms

Intro to Quant Trading Strategies (Lecture 7 of 10)
Intro to Quant Trading Strategies (Lecture 7 of 10)Intro to Quant Trading Strategies (Lecture 7 of 10)
Intro to Quant Trading Strategies (Lecture 7 of 10)Adrian Aley
 
Algorithm review
Algorithm reviewAlgorithm review
Algorithm reviewchidabdu
 
2010 3-24 cryptography stamatiou
2010 3-24 cryptography stamatiou2010 3-24 cryptography stamatiou
2010 3-24 cryptography stamatiouvafopoulos
 
Bias correction, and other uncertainty management techniques
Bias correction, and other uncertainty management techniquesBias correction, and other uncertainty management techniques
Bias correction, and other uncertainty management techniquesOlivier Teytaud
 
Uncertainties in large scale power systems
Uncertainties in large scale power systemsUncertainties in large scale power systems
Uncertainties in large scale power systemsOlivier Teytaud
 
constructing_generic_algorithms__ben_deane__cppcon_2020.pdf
constructing_generic_algorithms__ben_deane__cppcon_2020.pdfconstructing_generic_algorithms__ben_deane__cppcon_2020.pdf
constructing_generic_algorithms__ben_deane__cppcon_2020.pdfSayanSamanta39
 
Skiena algorithm 2007 lecture09 linear sorting
Skiena algorithm 2007 lecture09 linear sortingSkiena algorithm 2007 lecture09 linear sorting
Skiena algorithm 2007 lecture09 linear sortingzukun
 
Cheat Sheets for Hard Problems
Cheat Sheets for Hard ProblemsCheat Sheets for Hard Problems
Cheat Sheets for Hard ProblemsNeeldhara Misra
 
Fractal dimension versus Computational Complexity
Fractal dimension versus Computational ComplexityFractal dimension versus Computational Complexity
Fractal dimension versus Computational ComplexityHector Zenil
 
Parallelising Dynamic Programming
Parallelising Dynamic ProgrammingParallelising Dynamic Programming
Parallelising Dynamic ProgrammingRaphael Reitzig
 
Winter 8 Tutorial TM.pptx
Winter 8 Tutorial TM.pptxWinter 8 Tutorial TM.pptx
Winter 8 Tutorial TM.pptxHarisPrince
 
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppt
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppthanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppt
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.pptssuser148ae0
 
Skiena algorithm 2007 lecture08 quicksort
Skiena algorithm 2007 lecture08 quicksortSkiena algorithm 2007 lecture08 quicksort
Skiena algorithm 2007 lecture08 quicksortzukun
 
Is unit 4_number_theory
Is unit 4_number_theoryIs unit 4_number_theory
Is unit 4_number_theorySarthak Patel
 
Sienna 4 divideandconquer
Sienna 4 divideandconquerSienna 4 divideandconquer
Sienna 4 divideandconquerchidabdu
 

Similar a Noisy Optimization combining Bandits and Evolutionary Algorithms (20)

Intro to Quant Trading Strategies (Lecture 7 of 10)
Intro to Quant Trading Strategies (Lecture 7 of 10)Intro to Quant Trading Strategies (Lecture 7 of 10)
Intro to Quant Trading Strategies (Lecture 7 of 10)
 
Algorithm review
Algorithm reviewAlgorithm review
Algorithm review
 
2010 3-24 cryptography stamatiou
2010 3-24 cryptography stamatiou2010 3-24 cryptography stamatiou
2010 3-24 cryptography stamatiou
 
Bias correction, and other uncertainty management techniques
Bias correction, and other uncertainty management techniquesBias correction, and other uncertainty management techniques
Bias correction, and other uncertainty management techniques
 
Uncertainties in large scale power systems
Uncertainties in large scale power systemsUncertainties in large scale power systems
Uncertainties in large scale power systems
 
constructing_generic_algorithms__ben_deane__cppcon_2020.pdf
constructing_generic_algorithms__ben_deane__cppcon_2020.pdfconstructing_generic_algorithms__ben_deane__cppcon_2020.pdf
constructing_generic_algorithms__ben_deane__cppcon_2020.pdf
 
1. linear model, inference, prediction
1. linear model, inference, prediction1. linear model, inference, prediction
1. linear model, inference, prediction
 
Linear regression
Linear regressionLinear regression
Linear regression
 
Skiena algorithm 2007 lecture09 linear sorting
Skiena algorithm 2007 lecture09 linear sortingSkiena algorithm 2007 lecture09 linear sorting
Skiena algorithm 2007 lecture09 linear sorting
 
Cheat Sheets for Hard Problems
Cheat Sheets for Hard ProblemsCheat Sheets for Hard Problems
Cheat Sheets for Hard Problems
 
2019 PMED Spring Course - SMARTs-Part II - Eric Laber, April 10, 2019
2019 PMED Spring Course - SMARTs-Part II - Eric Laber, April 10, 2019 2019 PMED Spring Course - SMARTs-Part II - Eric Laber, April 10, 2019
2019 PMED Spring Course - SMARTs-Part II - Eric Laber, April 10, 2019
 
Fractal dimension versus Computational Complexity
Fractal dimension versus Computational ComplexityFractal dimension versus Computational Complexity
Fractal dimension versus Computational Complexity
 
A greedy algorithms
A greedy algorithmsA greedy algorithms
A greedy algorithms
 
Parallelising Dynamic Programming
Parallelising Dynamic ProgrammingParallelising Dynamic Programming
Parallelising Dynamic Programming
 
Winter 8 Tutorial TM.pptx
Winter 8 Tutorial TM.pptxWinter 8 Tutorial TM.pptx
Winter 8 Tutorial TM.pptx
 
36 greedy
36 greedy36 greedy
36 greedy
 
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppt
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppthanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppt
hanoosh-tictactoeqqqqqqqaaaaaaaaaaaa.ppt
 
Skiena algorithm 2007 lecture08 quicksort
Skiena algorithm 2007 lecture08 quicksortSkiena algorithm 2007 lecture08 quicksort
Skiena algorithm 2007 lecture08 quicksort
 
Is unit 4_number_theory
Is unit 4_number_theoryIs unit 4_number_theory
Is unit 4_number_theory
 
Sienna 4 divideandconquer
Sienna 4 divideandconquerSienna 4 divideandconquer
Sienna 4 divideandconquer
 

Último

Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessPixlogix Infotech
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?Igalia
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsJoaquim Jorge
 

Último (20)

Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Advantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your BusinessAdvantages of Hiring UIUX Design Service Providers for Your Business
Advantages of Hiring UIUX Design Service Providers for Your Business
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
Bajaj Allianz Life Insurance Company - Insurer Innovation Award 2024
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 

Noisy Optimization combining Bandits and Evolutionary Algorithms

  • 1. Bandit­based EDA for noisy optimization. P. Rolet, O. Teytaud Birmingham, 2009 Tao, Inria Saclay Ile-De-France, LRI (Université Paris Sud, France), UMR CNRS 8623, I&A team, Digiteo, Pascal Network of Excellence.
  • 2. Outline Introduction Lower bound An intuitive solution: bandits Rigorous upper-bound and rigorous solution Rolet and Teytaud TRSH 09 is great 2
  • 3. Introduction Noisy optimization: - the fitness function returns a noisy answer; - this noisy answer is independent; - goal: finding best expected value. Rolet and Teytaud TRSH 09 is great 3
  • 4. Main argument of this work Usual algorithms don't work in noisy optimization. The computational power is like this ==> <== and the result is like that because algos are not consistent. ==> there is much to win, algorithms can be greatly improved. Rolet and Teytaud TRSH 09 is great 4
  • 5. Previous works Should we average or not ? Or should we increase lambda ? ==> various answers in the literature. Rolet and Teytaud TRSH 09 is great 5
  • 6. Jebalia, Auger, PPSN 2008 . . ==> scale-invariant 1+1-ES converges linearly Rolet and Teytaud TRSH 09 is great 6
  • 7. Jebalia, Auger, PPSN 2008 . . In the present work: - no lower-bound on fnoise(x)/f(x) - “real” algorithm (not scale-invariant) ==> but slower rate (yet tight) Rolet and Teytaud TRSH 09 is great 7
  • 8. Introduction Introduction Lower bound An intuitive solution: bandits Rigorous upper-bound and rigorous solution Rolet and Teytaud TRSH 09 is great 8
  • 9. Lower bound: boring framework Optimization algorithm Rolet and Teytaud TRSH 09 is great 9
  • 10. Lower bound: boring framework Noisy (binary) measurement It's a lower bound. If it holds in the binary case, it holds in the non-binary case either. Rolet and Teytaud TRSH 09 is great 10
  • 11. Lower bound: boring framework Final loss Rolet and Teytaud TRSH 09 is great 11
  • 12. How to prove a lower bound ? Simple case... Consider we are in dimension 2. Consider that you solve the problem with precision . Consider a regular simplex of possible optima: d(ti,tj)= (here, for simplicity: deterministic algorithm) Rolet and Teytaud TRSH 09 is great 12
  • 13. How to prove a lower bound ? Simple case... Visit point x f(x,ti) = f(x,tj) ± so with proba 1-, f (x,ti) = f (x,tj) noise noise ==> with proba 1-, an iteration with optimum in ti = an iteration with optimum in tj Rolet and Teytaud TRSH 09 is great 13
  • 14. How to prove a lower bound ? Simple case... With proba 1-, an iteration with optimum in ti = an iteration with optimum in tj With proba 1-N, a run with optimum in ti = a run with optimum in tj ==> real case = similar to this one. Rolet and Teytaud TRSH 09 is great 14
  • 15. Lower bound Rolet and Teytaud TRSH 09 is great 15
  • 16. Introduction Introduction Lower bound An intuitive solution: bandits Rigorous upper-bound and rigorous solution Rolet and Teytaud TRSH 09 is great 16
  • 17. Idea of bandits I have N arms. I have t time steps. Pulling an arm yields a reward in [0,1]. Each arm has a stationary (i.i.d) reward. At each time step I can pull an arm. How will I find good arms ? Rolet and Teytaud TRSH 09 is great 17
  • 18. Idea of bandits The goal of Bernstein races: - guessing which arms are the most rewarding; - whilst saving up time. Rolet and Teytaud TRSH 09 is great 18
  • 19. Bernstein race: general idea. While (I want to go on) { I pull once each non discarded arm. I compute a lower and upper bound for all non-discarded arms (Bernstein bound). I discard arms which are excluded by the bounds. } Rolet and Teytaud TRSH 09 is great 19
  • 20. Already used for noisy optimization Idea: - evolutionary algorithm (CMA) - replacing selection by Bernstein race: keep racing until  points are selected. Trouble: - sometimes very expensive iterations - tricks are added in the algorithm, not stable Rolet and Teytaud TRSH 09 is great 20
  • 21. Our version Idea: - evolutionary algorithm; - derandomized mutation ensuring that there are at least two points with “sufficiently different” values. - replacing selection by Bernstein race: keep racing until  points are significantly better than 'other points. (=1, '=1) Tricky part: derandomized mutation (one more step in the algorithm; less simple); Otherwise no trouble, proved, we don't have to add tricks. Rolet and Teytaud TRSH 09 is great 21
  • 22. Introduction Introduction Lower bound An intuitive solution: bandits Rigorous upper-bound and rigorous solution Rolet and Teytaud TRSH 09 is great 22
  • 23. Bernstein race for comparing two among three arms with confidence 1-' Rolet and Teytaud TRSH 09 is great 23
  • 24. Bernstein race for comparing two among three arms with confidence 1-' Bernstein principle (low variance) Rolet and Teytaud TRSH 09 is great 24
  • 25. Bernstein race for comparing 2 among 3 arms with confidence 1-' THE important part: points are signif. different! Rolet and Teytaud TRSH 09 is great 25
  • 26. The complete algorithm The algorithm iteratively improves the set of possible optima. Iteration n as follows: 1) Generate 3 points (equally spaced on a line). 2) Apply the Bernstein race until one of the arms (the good arm) is statistically better than at least one other arm (the bad arm) for some ' = O(1/n2). 3) Remove the part of the domain which is farther from the good arm than from the bad arm. Sum of the ' =  ==> proof with confidence 1-. Rolet and Teytaud TRSH 09 is great 26
  • 27. The complete algorithm Rolet and Teytaud TRSH 09 is great 27
  • 28. Conclusion Bandits = good tool for noisy optimization. But taking care of how to generate points is necessary (otherwise one might have points with very similar fitness values which is dangerous!). Further work: * Generalizing the approach to more simple algorithms. * Lower bound = upper bound up to constant factors depending on the dimension * Proof of slower rate with variance not decreasing to zero at the optimum. Rolet and Teytaud TRSH 09 is great 28

Notas del editor

  1. I am Frederic Lemoine, PhD student at the University Paris Sud. I will present you my work on GenoQuery, a new querying module adapted to a functional genomics warehouse