UXPA Boston Conference 2018

Machine Learning (ML) is a mainstay science that is used to power personalization, recommendations, language translations, and other value additions to user experiences. Rather than relying on a set of logical defined relationships or hard-coded associations, ML algorithms learn from patterns in data without being manually programmed. It is changing the way we think and talk about product experiences. ML presents new opportunities for the designer, while adding a layer of complexity around user trust and control.

In this session, we unpack what machine learning is, what the core concepts of machine learning are and how it works. As we explore different kinds of learning techniques, we build a vocabulary that enables us, the design community, to engage with ML engineers and developers.

We don’t just stop there — we delve deeper into the gotchas of machine learning, specifically overfitting, fairness, bias and interpretability. We explore what these are, how they come into being, and why they matter. What do they mean for the design process? How do they emerge in the end user experience? How should we design user experiences that are no longer linear, and are instead a function of a complex ecosystem of interactions between users, interfaces, data, algorithms and environments?