Blind Federated Learning without initial model

Journal of Big Data

Table 6 Blended Blind FL results (constant weights, slope 5, sigmoid)—Breast Cancer dataset

Agent	Size	% 1s	Accuracy	Accuracy	Precision	Precision
Agent	Size	% 1s	Pre-FL	Post-FL	Pre-FL	Post-FL
1	20%	14%	0.6818	0.8636	0.3000	0.5000
2	20%	26%	0.8696	0.8696	0.7143	0.7143
3	20%	35%	0.7826	0.8261	0.6364	0.7000
4	20%	13%	0.8261	0.8261	0.4286	0.4286
5	20%	26%	0.9565	0.9565	0.8571	0.6667
Avg.	–	–	0.8233	0.8684	0.5873	0.6019
1	19%	24%	0.9063	0.9063	0.6667	0.6667
2	26%	35%	0.9365	0.9365	0.9200	0.8846
3	16%	42%	0.7073	0.7073	0.9231	0.9231
4	15%	20%	0.9487	0.9487	0.8571	0.7500
5	24%	47%	0.8704	0.8889	0.8696	0.9091
Avg.	–	–	0.8738	0.8775	0.8473	0.8267
1	5%	27%	1.0000	1.0000	1.0000	1.0000
2	4%	43%	0.9231	1.0000	0.7500	1.0000
3	42%	35%	0.9495	0.9495	0.9697	0.9697
4	20%	23%	0.9474	0.9474	0.7778	0.7778
5	29%	49%	0.8788	0.8788	1.0000	1.0000
Avg.	–	–	0.9397	0.9551	0.8995	0.9495
1	3%	20%	0.7778	0.8889	0.5000	0.5000
2	4%	36%	0.9091	0.9091	1.0000	1.0000
3	45%	40%	0.8785	0.8879	0.9375	0.9394
4	21%	17%	0.9524	0.9762	0.8333	1.0000
5	26%	49%	0.8833	0.9000	0.9310	0.9615
Avg.	–	–	0.8802	0.9124	0.8404	0.8802