Orateur
Description
The Application Driven Learning is a framework that integrates the predictive machine learning model training directly with the decision-making processes, optimizing predictions specifically for the application context.
We present ApplicationDrivenLearning.jl, a high-performance Julia package that enables efficient experimentation and implementation of the framework, particularly for large-scale decision-making problems. The package allows users to apply the novel gradient-based heuristic and the two original methods: the heuristic based on Nelder-Mead and Bilevel Optimization. Moreover, the heuristics have also been parallelized with MPI allowing the user to optimize their models in high-performance computing (HPC) clusters.
To demonstrate the usage of the package, we present a case study contrasting the multiple implementations that are available to the users.
