Child pages
  • Advanced Course

Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

But more importantly, we cover new topics, such as:

  • Parameter Sensitivity Analysis
  • Function Nodes
  • Influence Diagrams
  • Dynamic Bayesian Networks
  • Bayesian Updating
  • Aggregation of the Discrete States
  • Missing Values Processing
  • Credible/Confidence Intervals Analysis
  • Evidence Analysis
  • Function Optimization
  • Contribution Analysis

...

Localtab Group


Localtab
activetrue
titleModeling by Brainstorming

image2018-12-12_13-20-51.png

  • Expert-based Based Modeling via Brainstorming
  • Why Expert-based Based Modeling?
  • Value of Expert-based Based Modeling
  • Structural modelingModeling: Bottom-up Up and Top-down approachesDown Approaches
  • Parametric modelingModeling
  • Cognitive biasesBiases
  • BEKEE: Bayesia Expert Knowledge Elicitation Environment
    • Interactive
    • Batch
    • Segmentation of the Experts
    • Creation of Bayesian Belief Networks based on the elicited probabilitiesElicited Probabilities
    • Analysis of the Expert Assessments
    • Parameter Sensitivity Analysis
  • Experimentation of an Interactive session for probability elicitationExercise: Interactive Session for Probability Elicitation


Localtab
titleInfluence Diagrams


  • Utility Nodes
  • Decision Nodes
  • Expected Utility
  • Automatic Policy Optimization
  • Example: Oil Wildcatter
  • Exercices


Localtab
titleFunction Nodes


  • Motivations
  • Inference Functions
  • Formatting
  • Function Nodes as Parents
  • Exercise


Localtab
titleTemporal Dimension

  • Hidden Markov Chain
  • Unfolded Temporal Bayesian Networks
  • Dynamic Bayesian Networks
  • Temporal simulations Simulations (scenarios, temporal conditional dependencies, temporal monitoringScenarios, Temporal Conditional Dependencies, Temporal Monitoring)
  • Exact and approximate inferenceApproximate Inference
  • Unfolding Dynamic Bayesian Networks
  • Exercise: Maintenance of a Fluid Distribution System
  • Network Temporalization
  • Temporal Forecast
  • Exercise: Box & Jenkins


...

Localtab Group


Localtab
activetrue
titleBayesian Updating

Image Modified

  • Unrolled Networks
  • Compact Networks
    • Hyperparameters
    • Conditional dependenciesDependencies
  • Exercise: Bayesian Updating for Equine Anti-Doping


Localtab
titleDiscretization and Aggregation

  • Impact of Discretization
  • Requirements for a good discretizationGood Discretization
  • Pre and Post Discretization
  • Discretization viewed as the creation Creation of Latent variablesVariables
  • Discretization Methods
    • Manual by Expertise
    • Univariate
      • Equal Frequency
      • (Normalized) Equal Distance
      • Density Approximation
      • K-Means
      • R2-GenOpt
      • R2-GenOpt*
    • Bi-Variate
      • Tree
      • Perturbed Tree
    • Multi-Variate
      • Supervised with Random Forest
      • Unsupervised with Random Forest
      • R2-GenOpt
      • LogLoss-GenOpt
  • Exercise
  • Aggregation Methods for Symbolic variablesVariables
    • Manual by Expertise
    • Semi-Automatic
    • Bi-Variate with Tree
  • Exercise


Localtab
titleMissing Values


  • Types of Missingness:
    • Missing Completely at Random (MCAR)
    ,
    • Missing at Random (MAR)
    ,
    • Not Missing at Random (NMAR)
    ,
    • Filtered/
    censored
    • Censored/
    skipped
    • Skipped
  • Types of methodsMethods
    • Static
      • Filtering
      • A priori Priori Replacement
      • Entropy Based and Standard Static Imputation
    • Dynamic
      • Dynamic Imputation
      • Entropy Based Dynamic Imputation
      • Structural Expectation-Maximization
      • Approximate Dynamic Imputation with Static Imputation

  • Missing values imputation Values Imputation (Standard, Entropy-basedBased, Maximum Probable Explanation)

  • Exercise

  • Filtered/censoredCensored/skipped valuesSkipped Values

  • Example: Survey analysisAnalysis


Day 3

Localtab Group


Localtab
activetrue
titleVariable Synthesis


  • Manual Synthesis
  • Binarization
  • Clustering
    • K-Means
    • Bayesian Clustering
    • Hierarchical Bayesian Clustering
  • Exercises


Localtab
titleFine Tuning


  • Minimum Description Length (MDL) Score
  • Parameter Estimation with Trees
  • Structural Coefficient
  • Stratification
  • Smooth Probability Estimation
  • Exercise: CarStarts


Localtab
titleAnalysis


  • Credible/Confidence Interval Analysis
  • Evidence Analysis
    • Most Probable Explanation
    • Joint Probability of Evidence
    • Log-Loss
    • Information Gain
    • Bayes Factor
  • Performance Analysis
    • Supervised
    • Unsupervised
      • Compression
      • Multi-Target
  • Outlier Detection
  • Path Analysis
  • Exercises


Localtab
titleOptimization


  • Genetic algorithmAlgorithm
  • Objective Function
    • States/Mean
    • Function value
    • Maximization/Minimization
    • Target Value
    • Resources
    • Joint Probability/Support
  • Search Methods
    • Hard Evidence
    • Numerical Evidence
    • Direct Effects
  • Exercise: Marketing Mix Optimization


Localtab
titleContributions


  • Direct Effects
  • Type I Contribution
  • Type II Contribution
  • Base Mean
  • Normalization
  • Stacked Curves
  • Synergies


...