Mitigating Phone Fraud / What Are Transformers?




Mitigating Phone Fraud / What are transformers?

6 May 2019

New York

Added 01-Jan-1970

Speaker 1: Mitigating Phone Fraud with Machine Learning
Next Caller uses machine learning to assess the fraud risk of millions of phone calls per day for major financial institutions. Join Senior Architect César del Solar and Principal Data Scientist Jesse Day as they share their team's journey toward building an enterprise-grade machine learning infrastructure with AWS SageMaker from scratch.

Speaker 2:What are transformers? How can they be applied to banking?
Recent developments in sequence transduction models have improved natural language processing and understanding capabilities. In this talk, Art describes the transformer architecture and how it can be used to understand and evaluate symbolic variables and expressions embedded in plain text. With nearly perfect accuracy, such models can read text to understand operations and operands involving addition, subtraction and multiplication of both positive and negative decimal numbers of variable digits assigned to symbolic variables. Such sequence transduction models can be used to analyze financial reports, flagging suspicious or erroneous content.