Association·arcadia

Grid search tuning hyperparameters for training

Optimization of latent space size, hidden layer size, and noise level by grid search for dataset-specific model training.

Confidence
100%
active

Evidence Quote

For each dataset, we tuned the latent space size, hidden layer size, and noise amount using grid search.

Relationship

Model training procedure optimizes Model hyperparameters

Evidence

Reference discussing the influence of nonlinear signals on prediction performance

Heil BJ et al. (2022). The Effects of Nonlinear Signal on Expression-Based Prediction Performance doi:10.1101/2022.06.22.497194

Reference evaluating limitations of deep learning predictions for gene perturbation effects

Ahlmann-Eltze C et al. (2024). Deep learning-based predictions of gene perturbation effects do not yet outperform simple linear baselines doi:10.1101/2024.09.16.613342