Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures
Choosing a learning rate scheduling strategy is critical to neural network training, but manual selection is costly and rarely exhaustive. While classical AutoML approaches often treat the scheduler as a secondary hyperparameter, we systematically investigate its impact on classification accuracy across a diverse pool of architectures. We evaluated 30 representative architectures from convolutional and transformer families within the LEMUR neural network dataset. Through automated source-code injection, we applied 25 scheduler configurations across nine PyTorch families, evaluating a total of
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 47%fastai/fastai →
- PossiblePossibly related (embedding) · 46%sunrainyg/RandOpt →
- PossiblePossibly related (embedding) · 46%tensorflow/serving →
- PossiblePossibly related (embedding) · 45%deepspeedai/DeepSpeed →
- LinkedLinked via arxiv author · 85%Hafsa Mateen →
“Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures”
- LinkedLinked via arxiv author · 85%Radu Timofte →
“Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures”
- LinkedLinked via arxiv author · 85%Dmitry Ignatov →
“Systematic Evaluation of Learning Rate Scheduling Strategies Across Heterogeneous Architectures”