Applications of ML and Deep Learning in Financial Services

Almost every industry out there is being impacted by the application of Machine Learning and Deep Learning in some shape or form. Financial Services (FS) and Retail have epitomized some of the most cutting edge advancements in ML and Deep Learning techniques in recent years. From conversational bots to pricing optimization, to out of stock detection using computer vision, to agent-based modeling for portfolio management using reinforcement learning, and many more.

For this session, we have invited thought leaders in the Data Science space who apply their Machine Learning / Deep Learning expertise in either FS or Retail.

Speakers:

Paulo Rosairo, CFA, Senior Economist/Data Scientist at M&G Prudential

Topic: Using deep learning to identify profitable trades

More than just algorithms. Separating myth from the reality we will discuss a real-world example of how to use deep learning for investment decisions. Using a proprietary dataset of over 200,000 deals we leverage the power of new machine learning techniques to identify trades that are likely to outperform the market. Black box models might work, but as real money is on the line we will focus on understanding the reasons behind the decisions of the model.

Alejandro Saucedo, Chief Scientist, Institute of Ethical AI & Machine Learning

Alejandro is the Chief Scientist at the Institute for Ethical AI & Machine Learning, where he leads highly technical research on machine learning explainability, bias evaluation, reproducibility, and responsible design. With over 10 years of software development experience, Alejandro has held technical leadership positions across hyper-growth scale-ups and tech giants including Eigen Technologies, Bloomberg LP and Hack Partners. He has a strong track record building departments of machine learning engineers from scratch, and leading the delivery of large-scale machine learning system across the financial, insurance, legal, transport, manufacturing and construction sectors (in Europe, US and Latin America).

Topic: A practical guide towards explainability and bias evaluation in machine learning

In this talk, we demystify AI explainability through a practical hands-on case study. Our objective will be to automate a loan approval process by building and evaluating a deep learning model. We'll introduce motivations through the practical risks that arise with "undesired bias" & "black box models", and we will show tackle these challenges using tools from the latest research and domain knowledge.

Sushil Shah, Data Solution Architect, Microsoft

Topic: Using Anomaly Detection for Fraud Analytics

This talk demystifies the topic of Fraud Analytics and clarifies some of the fundamental choices that you have to make when constructing anomaly detection mechanisms. You'll learn why some approaches to anomaly detection work better than others in certain situations, and why a better solution for some challenges may be within reach after all. We will introduce Anomaly Detection in the context of Financial Fraud Analytics and what challenges can be addressed through these techniques. We will describe various Machine Learning / Deep Learning techniques for anomaly detection and will particularly emphasize Isolation Forests and Autoencoders.

Gary Short, Cloud Solution Architect (Advanced Analytics & AI), Microsoft

Topic: Turning Qualitative Data Into Quantitative Data Via NLP And Graphs: A Use Case

Handling qualitative data is a specialist niche in statistics and ML and not every analytics team in every organisation can do it. In this presentation I’ll tell you the story of one such large asset management company who were spending millions to collect qualitative data that then lay idle in the database as their team of quants didn’t know what to do with it. I’ll describe how I used NLP, CosmosDB, Graphviz and social network analysis tools in order to allow quants to exploit what was dormant qualitative data.