Adam-PFN, a prior-data fitted network trained on CDF-augmented real Adam learning curves, improves learning curve extrapolation and accelerates Adam hyperparameter tuning versus generic surrogates, with limited OOD gains.
During evaluation, we used the implementation suggested by Rakotoarison et al., 2024 and their publicly available surrogate model at https://github.com/automl/ifBO
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Tune My Adam, Please!
Adam-PFN, a prior-data fitted network trained on CDF-augmented real Adam learning curves, improves learning curve extrapolation and accelerates Adam hyperparameter tuning versus generic surrogates, with limited OOD gains.