BAI-OPTIMA

Major active project

BenAI common OPTimal Individualised decision MAking algorithms and foundational framework based on future intelligence envisaged by BenAI

BAI-OPTIMA originated from Clinical Decision Support System and has the ambition to be a foundational framework for decision making provided appropriate reasoning architecture be generalised and domain knowledge imported.

Major Innovations

See below

Maths

Decision logic (D-logic) extends traditional logic and probability.

Architecture

• The GID model: a prediction module + an abstraction system + a utility function + an abstraction-enforcement-module
• Enabling small dataset training and limited generalisation by introducing abstractions and inverse-abstractions

Algorithms

• The logic-MCUR algorithm: logic plus Multiple Criteria decision analysis within Uncertain Range
• DON (1-3): Discovery Of New abstractions
• The DON-I algorithm: tuning parameters of existing abstractions (rules)
• The DON-II algorithm: finding sufficient conditions
• The DON-III algorithm: model-agnostic probability based explainability
• The SIR algorithm: finding similar cases