Time: 2:00 pm – 3:00 pm
Location: Arora 16/17 Masterclass
This session is only bookable for those that have a confirmed place at Women in Data
This session has a limit of 130 spaces, please log in to your Women in Data account to register for a place.
Session title: Causality and Fairness in Machine Learning Projects
Speakers: Nisara Sriwattanaworachai and Dr Ines Marusic, Data Scientists, from QuantumBlack, Al by McKinsey
During this session Nisara will be presenting and discussing with the audience ‘Causality in machine Learning projects’. As we know, more and more organisations are using machine learning to help make business decisions. In this type of problem performance of models is no longer the only concern. In many machine learning projects, it is crucial to distinguish between events that cause outcomes and those that merely correlate. Nisara will discuss the concept of causality and how we use this to generate deeper insights for our clients.
Next, Viktoriia will be presenting on ‘Fairness in machine learning projects’. With an increasing number of machine learning projects in areas such as banking, finance, and pharmaceuticals, models are being applied to make decisions that can severely effect people’s lives. Unfortunately, the data in these domains often contains bias that exists in our society. The goal of algorithmic fairness is to ensure machine learning models make fair predictions devoid of discrimination. This presentation will show the full fairness pipeline from bias detection to evaluation of the model’s fairness to post-processing, to make the model’s predictions fairer using the example of a pharmaceutical case study.
During this workshop there will be opportunities for smaller problem solving discussions, on these topics, within the audience.