Transformers for Tabular Data at Capital One with Bayan Bruss - #591
Published:Sep 12, 2022 18:20
•1 min read
•Practical AI
Analysis
This article discusses the application of deep learning techniques, particularly transformers, to tabular data within the financial services sector, focusing on the work of Bayan Bruss at Capital One. It highlights the challenges of applying deep learning to tabular data and the opportunities presented by multi-modality and transformer models. The article also mentions research papers from Bruss's team on transformers and transfer learning for tabular data. The discussion touches upon the relative lack of attention given to tabular data research despite its widespread use in business.
Key Takeaways
- •The article focuses on applying deep learning, including transformers, to tabular data in financial services.
- •It highlights the challenges and opportunities related to using deep learning on tabular data.
- •The discussion includes research on transformers and transfer learning for tabular data from Bayan Bruss's team.
Reference
“We discuss why despite a “flood” of innovation in the field, work on tabular data doesn’t elicit as much fanfare despite its broad use across businesses...”