A Federated Learning-Inspired Evolutionary Algorithm: Application to Glucose Prediction

Abstract

In this paper, we propose an innovative Federated Learning-inspired evolutionary framework. Its main novelty is that this is the first time that an Evolutionary Algorithm is employed on its own to directly perform Federated Learning activity. A further novelty resides in the fact that, differently from the other Federated Learning frameworks in the literature, ours can efficiently deal at the same time with two relevant issues in Machine Learning, i.e., data privacy and interpretability of the solutions. Our framework consists of a master/slave approach in which each slave contains local data, protecting sensible private data, and exploits an evolutionary algorithm to generate prediction models. The master shares through the slaves the locally learned models that emerge on each slave. Sharing these local models results in global models. Being that data privacy and interpretability are very significant in the medical domain, the algorithm is tested to forecast future glucose values for diabetic patients by exploiting a Grammatical Evolution algorithm. The effectiveness of this knowledge-sharing process is assessed experimentally by comparing the proposed framework with another where no exchange of local models occurs. The results show that the performance of the proposed approach is better and demonstrate the validity of its sharing process for the emergence of local models for personal diabetes management, usable as efficient global models. When further subjects not involved in the learning process are considered, the models discovered by our framework show higher generalization capability than those achieved without knowledge sharing: the improvement provided by knowledge sharing is equal to about 3.03% for precision, 1.56% for recall, 3.17% for F1, and 1.56% for accuracy. Moreover, statistical analysis reveals the statistical superiority of model exchange with respect to the case of no exchange taking place.

Keywords: diabetes; evolutionary algorithms; federated learning; interpretable machine learning.

Figure 1

FLEA migration. The left side of the figure sketches the migration from the slaves to the master, while the right one traces the migration from the master to the slaves. The largest circles show single slaves with the corresponding local individuals, i.e., prediction models ($m_1,\dots ,m_n$). In particular, the individuals internal to the left-side circle have the same color because the models are all related only to local data. Differently, the individuals inside the right-side circle indicate that each slave, after communication with the master, can integrate the information from the immigrants by exploiting the mechanism of selection, replacement, and genetic operators, as evidenced by the color overlapping of the single individuals.

Figure 2

The grammar for the glucose forecasting model (Equation (5)).

Figure 3

Classification bands on the testing set for each patient in the Ohio data set [83].

Figure 4

Confusion matrices on the testing set of the best model (model 570) evolved by the non-FL approach on each patient.

Figure 4

Confusion matrices on the testing set of the best model (model 570) evolved by the non-FL approach on each patient.

Figure 5

Confusion matrices on the testing set of each patient for the best model (model 575) evolved by FLEA algorithm.

Figure 6

Confusion matrices on the testing set of new subjects not participating in the learning process achieved by the best model (model 570) evolved by non-FL approach.

Figure 7

Confusion matrices on the testing set of new subjects not participating in the learning process achieved by the best model (model 575) evolved by FLEA algorithm.