Blind Federated Learning without initial model

Journal of Big Data

Table 11 Blind FL results (Precision-based weights)—Breast Cancer dataset

Agent	Size	% 1s	Accuracy	Accuracy	Precision	Precision
Agent	Size	% 1s	Pre-FL	Post-FL	Pre-FL	Post-FL
1	20%	14%	0.7273	0.9091	0.3333	0.6000
2	20%	26%	0.8261	0.9130	0.6250	0.8333
3	20%	35%	0.7826	0.8696	0.6364	0.7778
4	20%	13%	0.8696	0.9130	0.5000	0.5000
5	20%	26%	0.9696	0.9565	0.6667	0.8571
Avg.	–	–	0.8150	0.9123	0.5523	0.7137
1	19%	24%	0.9318	0.9545	1.0000	0.8889
2	26%	35%	0.8667	0.8833	0.8421	0.8889
3	16%	42%	0.8889	0.8889	1.0000	1.0000
4	15%	20%	0.8824	0.8824	0.6364	0.6364
5	24%	47%	0.8727	0.8727	0.9524	0.9524
Avg.	–	–	0.8885	0.8964	0.8862	0.8733
1	5%	27%	0.6667	0.7500	0.4000	0.5000
2	4%	43%	0.8000	0.8000	1.0000	1.0000
3	42%	35%	0.9583	0.9688	1.0000	0.9091
4	20%	23%	0.9778	0.9778	0.9091	0.9091
5	29%	49%	0.7576	0.7879	0.9474	1.0000
Avg.	–	–	0.8321	0.8569	0.8513	0.8636
1	3%	20%	0.8333	0.8333	0.5000	0.5000
2	4%	36%	0.7000	0.8000	1.0000	1.0000
3	45%	40%	0.9423	0.9423	0.9500	0.9487
4	21%	17%	0.8958	0.9167	0.6364	0.5833
5	26%	49%	0.8689	0.8689	0.9600	0.9565
Avg.	–	–	0.8481	0.8722	0.8093	0.7977