BAI-OptiMa project

BAI-OptiMa (BenAI Optimal individualised decision Making foundational framework and algorithms) project

Aim of project: decision-making for critical areas like healthcare, including small dataset training and explainability

Major Innovations

Theory

  • D-logic (Decision logic) is a leap towards approaching decision-making method of human intelligence by incorporating hierarchical decision-making and domain/partial knowledge

Architecture

  • The GID ensemble model: enable small dataset training by introducing domain rules and trainable process that could extract domain expert experiences

Algorithms

  • The logic-MCUR algorithm: implementation of D-logic
  • The DoNde algorithm: AI explainability
  • The DeENa algorithm: extract domain data/rules from natural language corpus
  • The SiR algorithm: find similar cases
  • ODe algorithm: find optimal decision
  • TiSU algorithm: convert times series data into Gaussian
Scroll to Top