Columns represent the different hyperparameters. Each bar within a column represents a specific setting of a hyperparameter. The blue color indicates the number of successful configurations, while the red color represents the number of failed configurations. The vertical axis displays the distribution of failed configurations for each specific setting among the 6,480 tested configurations.</p
Number of disentangled data features for different VAE models having different latent dimensions.</p
<p>Overview of known and unknown parameters in alternative (reduced error) simulation scenarios.</p
<p>Negative values indicate that the model is less likely than the VVO model. Error bars represent s...
Columns represent the different hyperparameters. Each bar within a column represents a specific sett...
A listing of the hyperparameters that were held constant throughout the study. The values were set a...
The figure shows the WSEPIN score on the y−axis, while the bars are colored after the different late...
Each boxplot shows the clustering performance (ARI, x−axis) of fixing a hyperparameter while varying...
Scatter plot for the 90th percentile of the validation loss of different hyperparameters configurati...
Scatter plot for the 85th percentile of the validation loss of different hyperparameters configurati...
Each boxplot shows the clustering performance (ARI, x−axis) of fixing a hyperparameter while varying...
Plotting the 90th percentile (i.e., excluding the highest 10%) of the of the validation loss (y−axis...
<p>This figure has two kinds of bars, which respectively represent the failure rate on VM migration ...
Scatter plot for the 90th percentile of the validation loss of different hyperparameters configurati...
The 3.5 CV feature set was used and models produced using default and optimal hyperparameter setting...
A) Clustering performance (ARI, y−axis, the higher the better) in the latent space of each model (x−...
Number of disentangled data features for different VAE models having different latent dimensions.</p
<p>Overview of known and unknown parameters in alternative (reduced error) simulation scenarios.</p
<p>Negative values indicate that the model is less likely than the VVO model. Error bars represent s...
Columns represent the different hyperparameters. Each bar within a column represents a specific sett...
A listing of the hyperparameters that were held constant throughout the study. The values were set a...
The figure shows the WSEPIN score on the y−axis, while the bars are colored after the different late...
Each boxplot shows the clustering performance (ARI, x−axis) of fixing a hyperparameter while varying...
Scatter plot for the 90th percentile of the validation loss of different hyperparameters configurati...
Scatter plot for the 85th percentile of the validation loss of different hyperparameters configurati...
Each boxplot shows the clustering performance (ARI, x−axis) of fixing a hyperparameter while varying...
Plotting the 90th percentile (i.e., excluding the highest 10%) of the of the validation loss (y−axis...
<p>This figure has two kinds of bars, which respectively represent the failure rate on VM migration ...
Scatter plot for the 90th percentile of the validation loss of different hyperparameters configurati...
The 3.5 CV feature set was used and models produced using default and optimal hyperparameter setting...
A) Clustering performance (ARI, y−axis, the higher the better) in the latent space of each model (x−...
Number of disentangled data features for different VAE models having different latent dimensions.</p
<p>Overview of known and unknown parameters in alternative (reduced error) simulation scenarios.</p
<p>Negative values indicate that the model is less likely than the VVO model. Error bars represent s...