Edinburgh Napier University

  The project proposes a novel architecture for designing optimisation systems that “keeps learning” when exposed to a continual stream of instances: these systems will autonomously adapt their components over time to respond to gradual drift in the characteristics of instances faced and also predict characteristics of future instances, enabling them to prepare by generating suitable solvers in advance. The proposal is supported by three collaborating companies who each develop optimisation software to tackle streams of instances faced in dynamic and fast-changing environments.keep on learning