What type of study is this?

September 10, 2025Mathematical Modeling and Computing

The impact of activation functions on LTSM server load prediction accuracy: machine learning approach

View Full Paper

Ask AI

Bookmark

Authors

ОЧОлеся Чабан PPPavlo Pukach PPPavlo Pukach

Discussion

Member takes

Implication

Observational analysis shows differences in prediction accuracy across activation functions in LSTM models, highlighting their significance for server load forecasting.

Key Points

Swish and ReLU activation functions demonstrated lower mean squared error values compared to the baseline.
The study found a reduction in prediction errors ranging from 6.6–16.3% with optimal activation function choices.
Activation function selection significantly influences model stability and predictive reliability in LSTM-based analyses.
The sigmoid function misled accuracy metrics, underestimating peak loads and showing poor predictive fidelity.

Cite This Study

Чабан et al. (2025) studied this question.

synapsesocial.com/papers/68c18f469b7b07f3a061642e https://doi.org/10.23939/mmc2025.03.779

View Full Paper

Ask AI

Bookmark

Implication

Observational analysis shows differences in prediction accuracy across activation functions in LSTM models, highlighting their significance for server load forecasting.

The impact of activation functions on LTSM server load prediction accuracy: machine learning approach

Discussion

Member takes

Implication

Key Points

Cite This Study

The impact of activation functions on LTSM server load prediction accuracy: machine learning approach

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider