Minimum Description Length (MDL)
GPTKB entity
Statements (49)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
gptkb:statistical_principle
|
| gptkbp:abbreviation |
Minimum Description Length
|
| gptkbp:appliesTo |
data compression algorithms
parametric models model selection in statistics non-parametric models |
| gptkbp:approach |
formalization of Occam's razor
|
| gptkbp:criteria |
penalized likelihood
model complexity regularization |
| gptkbp:defines |
The best model is the one that minimizes the total description length of the model and the data given the model.
|
| gptkbp:field |
gptkb:information_theory
gptkb:machine_learning statistics |
| gptkbp:hasApplication |
gptkb:dictionary
gptkb:signal_processing natural language processing time series analysis bioinformatics network analysis regression clustering feature selection data mining image analysis anomaly detection pattern discovery graph mining |
| gptkbp:hasConcept |
best explanation is the one that leads to the best compression of the data
|
| gptkbp:influenced |
Bayesian inference
pattern recognition data mining model selection criteria |
| gptkbp:influencedBy |
gptkb:Shannon's_information_theory
algorithmic information theory |
| gptkbp:proposedBy |
gptkb:Jorma_Rissanen
|
| gptkbp:relatedConcept |
gptkb:Kolmogorov_complexity
gptkb:Minimum_Message_Length_(MML) gptkb:Occam's_razor |
| gptkbp:relatedTo |
gptkb:Minimum_Message_Length_(MML)
Akaike Information Criterion (AIC) Bayesian Information Criterion (BIC) |
| gptkbp:usedFor |
data compression
statistical inference model selection |
| gptkbp:website |
https://www.mdl-research.org/
|
| gptkbp:yearProposed |
1978
|
| gptkbp:bfsParent |
gptkb:Minimum_Message_Length_(MML)
|
| gptkbp:bfsLayer |
7
|
| https://www.w3.org/2000/01/rdf-schema#label |
Minimum Description Length (MDL)
|