Table 7 Statistical significance analysis for prediction performance (p values from paired t test).

From: Attention-based workload prediction and dynamic resource allocation for heterogeneous computing environments

Comparison

MAE

RMSE

MAPE

R² Score

Proposed versus LSTM

< 0.001***

< 0.001***

< 0.001***

< 0.001***

Proposed versus GRU

< 0.001***

< 0.001***

< 0.001***

< 0.001***

Proposed versus vanilla transformer

< 0.001***

< 0.001***

< 0.001***

< 0.001***

Proposed versus informer

0.003**

0.005**

0.004**

0.008**

Proposed versus autoformer

0.008**

0.012*

0.009**

0.015*

Proposed versus FEDformer

0.021*

0.028*

0.024*

0.031*

Proposed versus PatchTST

0.042*

0.056

0.048*

0.063

  1. ***p < 0.001, ** p < 0.01, *p < 0.05; Bonferroni-corrected thresholds applied.
Back to article page