Salvato in:
Dettagli Bibliografici
Autori principali: Joseph, Manu, Raj, Harsh
Natura: Preprint
Pubblicazione: 2022
Soggetti:
Accesso online:https://arxiv.org/abs/2207.08548
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916086310502400
author Joseph, Manu
Raj, Harsh
author_facet Joseph, Manu
Raj, Harsh
contents We propose a novel high-performance, interpretable, and parameter \& computationally efficient deep learning architecture for tabular data, Gated Adaptive Network for Deep Automated Learning of Features (GANDALF). GANDALF relies on a new tabular processing unit with a gating mechanism and in-built feature selection called Gated Feature Learning Unit (GFLU) as a feature representation learning unit. We demonstrate that GANDALF outperforms or stays at-par with SOTA approaches like XGBoost, SAINT, FT-Transformers, etc. by experiments on multiple established public benchmarks. We have made available the code at github.com/manujosephv/pytorch_tabular under MIT License.
format Preprint
id arxiv_https___arxiv_org_abs_2207_08548
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle GANDALF: Gated Adaptive Network for Deep Automated Learning of Features
Joseph, Manu
Raj, Harsh
Machine Learning
We propose a novel high-performance, interpretable, and parameter \& computationally efficient deep learning architecture for tabular data, Gated Adaptive Network for Deep Automated Learning of Features (GANDALF). GANDALF relies on a new tabular processing unit with a gating mechanism and in-built feature selection called Gated Feature Learning Unit (GFLU) as a feature representation learning unit. We demonstrate that GANDALF outperforms or stays at-par with SOTA approaches like XGBoost, SAINT, FT-Transformers, etc. by experiments on multiple established public benchmarks. We have made available the code at github.com/manujosephv/pytorch_tabular under MIT License.
title GANDALF: Gated Adaptive Network for Deep Automated Learning of Features
topic Machine Learning
url https://arxiv.org/abs/2207.08548