Skip to main content

Table 9 % Accuracy (average ± standard deviation) obtained for each algorithm and source using DeepSleepNet

From: Comparison of deep transfer learning algorithms and transferability measures for wearable sleep staging

 

CiCC

ISRUC

MASS

SHHS

WSC

MrOS

Head Re-train

72.4 ± 9.6

59.9 ± 9.0

63.2 ± 9.4

62.3 ± 8.9

59.9 ± 8.4

64.2 ± 10.9

Subspace alignment

62.0 ± 6.8

55.3 ± 9.6

53.1 ± 11.4

52.4 ± 8.2

50.8 ± 9.2

53.5 ± 11.4

CORAL

70.6 ± 9.3

60.5 ± 8.8

61.8 ± 9.6

60.3 ± 8.1

58.8 ± 7.4

62.9 ± 8.8

Per-Class CORAL

70.8 ± 8.5

57.0 ± 8.8

62.0 ± 8.8

59.8 ± 7.4

57.7 ± 9.2

60.6 ± 10.8

DDC

N/A

N/A

N/A

N/A

N/A

N/A