Table 1:

Performance of Several models trained using classical machine learning methods, and federated learning methods, where the number of participating clients in the federation is $N=2$, tested on the PPMI dataset. Data reported is mean and standard deviation across K=6 fold cross validation.

Algorithm Name	ROC-AUC	AUC-PR	Balanced Accuracy	Precision	Recall	F 0.5	F 1	F 2	Log Loss	Matthews Correlation Coefficient
AdaBoost Classifier	0.865±0.033	0.939±0.01	0.736±0.069	0.844±0.043	0.89±0.047	0.897±0.017	0.885±0.027	0.942±0.013	0.632±0.007	0.499±0.122
Bagging Classifier	0.82±0.05	0.916±0.025	0.69±0.052	0.813±0.027	0.897±0.061	0.865±0.021	0.862±0.024	0.933±0.01	1.001±0.382	0.428±0.123
GradientBoosting Classifier	0.879±0.046	0.943±0.026	0.723±0.072	0.833±0.042	0.911±0.021	0.916±0.03	0.894±0.022	0.942±0.015	0.444±0.099	0.486±0.119
KNeighbors Classifier	0.61±0.099	0.806±0.065	0.533±0.029	0.729±0.014	0.937±0.046	0.782±0.023	0.837±0.007	0.927±0.004	2.836±0.617	0.111±0.104
LinearDiscriminantAnalysis Classifier	0.763±0.045	0.883±0.031	0.681±0.053	0.826±0.04	0.77±0.05	0.7±0.344	0.714±0.35	0.776±0.38	1.608±0.488	0.347±0.095
LogisticRegression Classifier	0.831±0.068	0.915±0.039	0.734±0.072	0.841±0.043	0.894±0.028	0.872±0.047	0.883±0.033	0.939±0.011	0.648±0.203	0.493±0.134
MLP Classifier	0.739±0.078	0.892±0.032	0.703±0.059	0.833±0.038	0.815±0.054	0.843±0.038	0.858±0.034	0.932±0.013	6.616±1.844	0.402±0.119
QuadraticDiscriminantAnalysis Classifier	0.504±0.057	0.774±0.029	0.504±0.057	0.725±0.055	0.385±0.081	0.757±0.008	0.833±0.006	0.926±0.003	19.674±1.492	0.009±0.105
RandomForest	0.816±0.076	0.917±0.027	0.552±0.034	0.736±0.016	0.993±0.017	0.857±0.043	0.874±0.032	0.942±0.014	0.508±0.029	0.249±0.121
SGD Classifier	0.755±0.065	0.907±0.025	0.735±0.062	0.846±0.032	0.857±0.068	0.857±0.037	0.876±0.036	0.936±0.015	7.525±2.282	0.481±0.143
SVC Classifier	0.838±0.069	0.924±0.032	0.711±0.071	0.827±0.041	0.883±0.042	0.872±0.042	0.886±0.024	0.941±0.008	0.44±0.082	0.447±0.145
XGBoost Classifier	0.89±0.046	0.953±0.018	0.765±0.097	0.86±0.062	0.911±0.03	0.915±0.03	0.900±0.033	0.942±0.014	0.461±0.135	0.557±0.167
XGBoost Random Forest Classifier	0.857±0.064	0.936±0.029	0.773±0.057	0.868±0.04	0.885±0.047	0.907±0.039	0.891±0.041	0.936±0.011	1.79±0.853	0.558±0.105
FedAvg LR	0.69±0.16	0.874±0.042	0.617±0.109	0.772±0.054	0.955±0.037	0.818±0.054	0.863±0.026	0.935±0.008	0.655±0.14	0.278±0.25
FedAvg MLP	0.76±0.102	0.872±0.072	0.671±0.087	0.817±0.051	0.768±0.089	0.708±0.35	0.728±0.358	0.779±0.382	0.767±0.308	0.334±0.179
FedAvg SGD	0.828±0.048	0.92±0.025	0.757±0.048	0.904±0.049	0.707±0.033	0.871±0.032	0.872±0.018	0.939±0.008	0.545±0.032	0.47±0.084
FedAvg XGBRF	0.829±0.023	0.924±0.015	0.739±0.058	0.848±0.043	0.883±0.036	0.886±0.02	0.875±0.012	0.929±0.005	0.691±0.0	0.497±0.089w
FedProx μ = 0.5 LR	0.755±0.142	0.887±0.041	0.653±0.088	0.791±0.042	0.941±0.031	0.704±0.349	0.729±0.358	0.784±0.384	0.609±0.155	0.362±0.198
FedProx μ = 0.5 MLP	0.757±0.096	0.872±0.061	0.695±0.088	0.829±0.048	0.808±0.075	0.843±0.042	0.868±0.028	0.937±0.004	0.976±0.314	0.387±0.182
FedProx μ = 2 LR	0.812±0.079	0.906±0.04	0.658±0.028	0.79±0.014	0.937±0.025	0.866±0.045	0.879±0.025	0.941±0.006	0.582±0.137	0.398±0.069
FedProx μ = 2 MLP	0.765±0.079	0.868±0.06	0.694±0.069	0.83±0.042	0.798±0.045	0.706±0.348	0.724±0.355	0.781±0.382	0.9±0.368	0.379±0.133