This technique ensures that the criteria of selection focuses on redundant filters, while retaining the rare ones, thus maximizing the variety of remaining filters.
Neural networks usually involve a large number of parameters, which correspond to the weights of the network.
Most of the time, sparsity is introduced using a three-stage pipeline: 1) train the model to convergence, 2) prune the model according to some criterion, 3) fine-tune the pruned model to recover performance.
Parameters are estimated by comparing the predictions of the metamodel with real data obtained from sensors using the CMA-ES algorithm, a derivative free optimization procedure.
Finally, the optimal settings to minimize the energy loads while maintaining a target thermal comfort and air quality are obtained using a multi-objective optimization procedure.