Why do tree-based models still outperform deep learning on tabular data? (2022) (arxiv.org)

Screenshot of arxiv.org: Why do tree-based models still outperform deep learning on tabular data? (2022)

First screen, captured 2 years ago.

Looks similar

Screenshot of arxiv.org: Deep Learning Is Not So Mysterious or Different

Deep Learning Is Not So Mysterious or Different (arxiv.org)

445 points by wuubuu 1 year ago | 112 comments

arXiv paper by Andrew Gordon Wilson arguing deep learning generalization phenomena like benign overfitting and double descent are explainable by soft inductive biases.

deep-learning generalization double-descent pac-bayes

Screenshot of arxiv.org: Why do random forests work? They are self-regularizing adaptive smoothers

Why do random forests work? They are self-regularizing adaptive smoothers (arxiv.org)

267 points by sebg 2 years ago | 37 comments

Screenshot of arxiv.org: Value-Based Deep RL Scales Predictably

Value-Based Deep RL Scales Predictably (arxiv.org)

64 points by bearseascape 1 year ago | 2 comments

Screenshot of arxiv.org: Trees to Flows and Back: Unifying Decision Trees and Diffusion Models

Trees to Flows and Back: Unifying Decision Trees and Diffusion Models (arxiv.org)

44 points by rsn243 4 months ago | 8 comments

Screenshot of arxiv.org: Trinary Decision Trees for missing value handling

Trinary Decision Trees for missing value handling (arxiv.org)

41 points by PaulHoule 3 years ago | 18 comments

Screenshot of arxiv.org: There Will Be a Scientific Theory of Deep Learning

There Will Be a Scientific Theory of Deep Learning (arxiv.org)

320 points by jamie-simon 6 months ago | 142 comments

An arXiv position paper arguing that a scientific theory of deep learning is emerging, surveying five strands of research such as solvable models, scaling limits, and universal behaviors.

deep-learning-theory deep-learning neural-networks machine-learning

Looks similar

Deep Learning Is Not So Mysterious or Different(arxiv.org)

Why do random forests work? They are self-regularizing adaptive smoothers(arxiv.org)

Value-Based Deep RL Scales Predictably(arxiv.org)

Trees to Flows and Back: Unifying Decision Trees and Diffusion Models(arxiv.org)

Trinary Decision Trees for missing value handling(arxiv.org)

There Will Be a Scientific Theory of Deep Learning(arxiv.org)