Stay organized with collections
Save and categorize content based on your preferences.
This page lists minor engine versions available to all
AML AI customers. For more information on engine version naming, see
the
engine version glossary entry.
aml-retail.default
This table lists the retail-specific engine versions.
Tuning
Major
Minor
Description
Release date
Lifecycle stage
v004
004
Latest version:
202408-000
Reduction in total requirement for transaction and account data from 41 to 30 months
Performance improvements across several feature families, focusing on more recent high risk activity
Adjustment to the calculation of the PartyRecall metric in the rare corner case when many customers have the same prediction score and it's not possible to exactly produce the number of partyInvestigationsPerPeriod positive predictions
Uses latest FATF high risk geos, published in Jan 2024
September 16, 2024
ACTIVE
002
202406-000
Additional data validation errors with more granular checks and corresponding actionable error messages
Improved accuracy and better descriptions for existing data validation checks
A bugfix for processing of alert events in the Risk Case Event table
Improved reliability of training, prediction and backtesting operations for very large datasets, >> 20 million parties
Reduction in time taken for tuning, that is creating an engine config
July 11, 2024
ACTIVE
001
202402-000
More sensitive skew metric for model and data quality monitoring.
Bugfix for recall metrics.
April 25, 2024
ACTIVE
000
202401-000
Adds feature family Importance metric to the model metadata
March 12, 2024
ACTIVE
202312-001
Improves usability of party supplementary data by allowing use of human readable party supplementary data IDs
Adds support for inheriting hyperparameters instead of re-tuning
March 1, 2024
DECOMMISSIONED
202312-000
Supports datasets containing up to 130 million parties
Adds KYC and tenure features versus the previous version
December 1, 2023
LIMITED
v003
000
202312-000
Improves usability of party supplementary data by allowing use of human readable party supplementary data IDs
Adds support for inheriting hyperparameters instead of re-tuning
March 1, 2024
ACTIVE
202311-000
Supports datasets containing up to 20 million parties
V003 adds improved labeling methodology versus previous versions
November 15, 2023
LIMITED
aml-commercial.default
This table lists the commercial-specific engine versions.
Tuning
Major
Minor
Description
Release date
Lifecycle stage
v004
004
Latest version:
202408-000
Reduction in total requirement for transaction and account data from 41 to 30 months
Performance improvements across several feature families, focusing on more recent high risk activity
Adjustment to the calculation of the PartyRecall metric in the rare corner case when many customers have the same prediction score and it's not possible to exactly produce the number of partyInvestigationsPerPeriod positive predictions
Uses latest FATF high risk geos, published in Jan 2024
September 16, 2024
ACTIVE
002
202406-000
Additional data validation errors with more granular checks and corresponding actionable error messages
Improved accuracy and better descriptions for existing data validation checks
A bugfix for processing of alert events in the Risk Case Event table
Improved reliability of training, prediction and backtesting operations for very large datasets, >> 20 million parties
Reduction in time taken for tuning, that is creating an engine config
July 11, 2024
ACTIVE
001
202402-000
More sensitive skew metric for model and data quality monitoring.
Bugfix for recall metrics.
April 25, 2024
ACTIVE
000
202401-001
More reliable tuning performance, in particular for small datasets
March 13, 2024
ACTIVE
202401-000
Adds feature family Importance metric to the model metadata
March 12, 2024
DECOMMISSIONED
202312-001
Improves usability of party supplementary data by allowing use of human readable party supplementary data IDs
Adds support for inheriting hyperparameters instead of re-tuning
March 1, 2024
DECOMMISSIONED
202312-000
Supports datasets containing up to 130 million parties
Performance neutral versus the previous version
December 1, 2023
LIMITED
v003
000
202312-000
Improves usability of party supplementary data by allowing use of human readable party supplementary data IDs
Adds support for inheriting hyperparameters instead of re-tuning
March 1, 2024
ACTIVE
202311-000
Supports datasets containing up to 20 million parties
V003 adds improved labeling methodology versus previous versions