Table 2 Human activity recognition, per time-step classification

From: Closed-form continuous-time neural networks

Model

Accuracy (%)

Time per epoch (min)

†RNN-Impute7

79.50 ± 0.8

0.38

†RNN-Δt7

79.50 ± 0.8

0.45

†RNN-Decay7

80.00 ± 1.0

0.39

†GRU-D51

80.60 ± 0.7

0.15

†RNN-VAE7

34.30 ± 4.0

2.63

†Latent-ODE-RNN7

83.50 ± 1.0

7.71

†ODE-RNN7

82.90 ± 1.6

3.15

†Latent-ODE-ODE7

84.60 ± 1.3

8.49

Cf-S (current work)

87.04 ± 0.47

0.097

CfC-noGate (current work)

85.57 ± 0.34

0.093

CfC (current work)

84.87 ± 0.42

0.084

CfC-mmRNN (current work)

85.97 ± 0.25

0.128

  1. Numbers represent mean ± s.d. (n = 5). The performance of the models marked by † is reported from ref. 7. Bold values indicate the highest accuracy and best time per epoch (min).
Back to article page