Departmental Seminar - Multimodal Tabular Transactions | Dr. Natraj Raman
| Dates: | 13 November 2024 |
|---|
| Times: | 14:00 - 15:00 |
|---|
| What is it: | Seminar |
|---|
| Organiser: | Department of Computer Science |
|---|
| How much: | Free |
|---|
| Who is it for: | University staff |
|---|
| Speaker: | Dr Natraj Raman |
|---|
|
Large Language Models (LLMs) have achieved remarkable success in understanding and reasoning over unstructured text data. Does this success suggest that structured formats, like tabular data, should also be converted into text for LLM application? This talk will delve into the advantages of adapting transformer-based models to the unique characteristics of tabular data, with a focus on financial transactions. It will also discuss scalable solutions to the practical challenges encountered in industrial settings for learning and interfacing with transaction data representations.
Speaker
Dr Natraj Raman
Role: Research Director
Organisation: JPMorgan AI Research, London
Biography: Natraj Raman is a research director at JPMorgan AI Research, London. With over 20 years of experience in the financial industry, he has worked as a research scientist at Thomson Reuters and S&P Global. He holds a PhD in Computer Vision, and his research interests include Document AI, Explainable AI and Representation Learning.
Travel and Contact Information
Find event
Kilburn_TH 1.3
Kilburn Building
Manchester
