Benchmarking Optimizers for MLPs in Tabular Deep Learning

Abstract

MLP is a heavily used backbone in modern deep learning (DL) architectures for supervised learning on tabular data, and AdamW is the go-to optimizer used to train tabular DL models. Unlike architecture design, however, the choice of optimizer for tabular DL has not been examined systematically, despite new optimizers showing promise in other domains. To fill this gap, we benchmark \Noptimizers optimizers on \Ndatasets tabular datasets for training MLP-based models in the standard supervised learning setting under a shared experiment protocol. Our main finding is that the Muon optimizer consistently outperforms AdamW, and thus should be considered a strong and practical choice for practitioners and researchers, if the associated training efficiency overhead is affordable. Additionally, we find exponential moving average of model weights to be a simple yet effective technique that improves AdamW on vanilla MLPs, though its effect is less consistent across model variants.

Workflow Status

Review status

pending

Role

unreviewed

Read priority

later

Vote

Not set.

Saved

no

Collections

Not filed yet.

Next action

Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

Similar Papers

BibTeX

@article{gorishniy2026benchmarking,
  title = {Benchmarking Optimizers for MLPs in Tabular Deep Learning},
  author = {Yury Gorishniy and Ivan Rubachev and Dmitrii Feoktistov and Artem Babenko},
  year = {2026},
  abstract = {MLP is a heavily used backbone in modern deep learning (DL) architectures for supervised learning on tabular data, and AdamW is the go-to optimizer used to train tabular DL models. Unlike architecture design, however, the choice of optimizer for tabular DL has not been examined systematically, despite new optimizers showing promise in other domains. To fill this gap, we benchmark \textbackslash{}Noptimizers optimizers on \textbackslash{}Ndatasets tabular datasets for training MLP-based models in the standard supervised learn},
  url = {https://arxiv.org/abs/2604.15297},
  keywords = {cs.LG, Benchmarking, Benchmark (surveying), Computer science, Artificial intelligence, Deep learning},
  eprint = {2604.15297},
  archiveprefix = {arXiv},
}

Metadata

{}