Trustworthiness and SDFL#62

Merged

enriquetomasmb merged 67 commits into

CyberDataLab:mainfrom

Jun 12, 2026

Commits on Mar 4, 2026

DFL trustworthiness and numerical datasets
juanjetm
committed

Commits on Mar 6, 2026

Minor changes in trustworthiness
juanjetm
committed

Commits on Mar 11, 2026

KDDCUP99 dataset added and errors fixed
juanjetm
committed

Commits on Mar 12, 2026

Commits on Mar 13, 2026

All CFL files updated
juanjetm
committed

Commits on Mar 16, 2026

CFL totally adapted
juanjetm
committed

Commits on Mar 17, 2026

Desynchronization error in CFL, accuracy/loss error and reports fixed
juanjetm
committed

Commits on Mar 18, 2026

Expected nodes fixed
juanjetm
committed

Commits on Mar 25, 2026

Commits on Apr 8, 2026

Commits on Apr 10, 2026

Privacy metrics implemented: MIA AUC and Epsilon Star, all new metrics added to eval_metrics and trustworthiness calculation, new notion added: outcome_fairness, other minor changes
juanjetm
committed

Commits on Apr 14, 2026

Some metrics changed and fixed, and macro F1, Reputation Enabled and Average Reputation added, other minor changes
juanjetm
committed

Commits on Apr 15, 2026

Commits on Apr 17, 2026

Commits on Apr 20, 2026

Commits on Apr 24, 2026

Commits on Apr 27, 2026

Commits on Apr 28, 2026

Commits on Apr 29, 2026

Commits on Apr 30, 2026

DP error fixed DFL, calculation refactoring, maxgradnorm added to frontend, DP error fixed CFL (always True)
juanjetm
committed

Commits on May 4, 2026

Factsheet refactoring, CFL reputation metrics added, CFL reputation updated and fixed
juanjetm
committed

Commits on May 5, 2026

SDFL Fixed: Forwarding trainer and aggregator, aggregation fixed, new message, leadership/ACK fixed
juanjetm
committed

Commits on May 6, 2026

Commits on May 11, 2026

Reputation implemented for SDFL. Leadership updated.
juanjetm
committed

Commits on May 14, 2026

Mitigation: Feature Squeezing for images
juanjetm
committed

Commits on May 15, 2026

Commits on May 19, 2026

Adversarial Training implemented for images, trustworthiness: FGSM fixed for images, frontend updated (differential privacy, feature squeezing, adversarial training)
juanjetm
committed

Commits on May 20, 2026

Models updated: Data type, tabular datasets fixed and continous/binary division, tabular attack configuration updated: Sample poisoning not available
juanjetm
committed

Commits on May 21, 2026

Commits on May 25, 2026

Commits on May 26, 2026

KDDCUP fixed, adversarial training working for KDD
juanjetm
committed

Commits on May 27, 2026

Global privacy risk fixed. Important SDFL refactoring.
juanjetm
committed

Commits on May 28, 2026

Adversarial training updated for tabular data: CAA V1 implemented
juanjetm
committed

Commits on May 29, 2026

Commits on Jun 1, 2026

Refactoring scoring, trust_reports, scenario_metrics, factsheet_values, and many functionality commented, other changes
juanjetm
committed

Commits on Jun 3, 2026

Commits on Jun 4, 2026

Adversarial Training CAPGD V1
juanjetm
committed

Commits on Jun 5, 2026

Commits on Jun 8, 2026

Adversarial training for tabular data updated: Margin window implemented. Covtype and KDDCUP99 updated and fixed for adversarial training
juanjetm
committed

Commits on Jun 10, 2026

Robustness metrics revised and fixed, trustworthiness metrics revised, models updated: Data type, factsheets updated, tabular robustness metrics deleted
juanjetm
committed

Commits on Jun 11, 2026

Commits on Jun 12, 2026