At Deeploy we find it crucial to make AI as transparent and explainable as possible. Although algorithms can often be difficult or even impossible to explain to people, its dynamics and predictions can be translated to human understandable explanations.

By making models explainable, we can keep better control on models, detect (ethical) risks and improve performance of models. Despite its relevance and importance, Explainability in AI (XAI) is not yet widely used in practice.

Deeploy is a platform for the deployment of models and explainers. It's built with data scientists in mind, giving autonomy to the data scientist to iterate over their models.

In this workshop, we will give a brief overview of the most important concepts, techniques, and the latest developments in the field of XAI. With this knowledge, you’re ready to get your first hands-on experience. Using design thinking methodologies, you will come up with solutions for a real-life explainability problem from the FinTech industry.

In this online workshop

• You will learn about the foundations of XAI

• You will get introduced to different XAI techniques

• You will practice reframing ML problems to user problems

• You will come up with user-centered explainability solutions for a real-world case, and learn how to do the same for problems in your own organization

The workshop leaders:

Sophia Zitman is an XAI R&D Engineer at Deeploy. She enjoys combining the pragmatic side of consulting with new research in academia. This unique position allows her to build state of the art applications in real life. Designing human centred tech is her strong suit.

Lara Zijlstra is a designer with a passion for digital products that lie at the intersection between users and ML. She currently works as a product designer at Deeploy, with previous experience designing AI-powered solutions at ING WB Advanced Analytics. With an academic background in business informatics, she is eager to build solutions that are the complete package; user friendly, technically feasible, and viable in the real world.

General information:

Study load: We will send you a 2-pager brief on the case that we will work on during the workshop, so you can get familiar with it beforehand.

For whom: This workshop is for anyone working in FinTech, who wants to learn how to start bringing explainable AI into practice.

Tools: The workshop will be online thru Google Meet and we will also use Mural for collaboration (no installation required).

About Deeploy:

The journey of Deeploy started in 2020, but the ideas that ended up being the foundation for Deeploy were shaped over many years. With a growing number of ML models going to production, we believed it is time for a solution that ensures manageable, explainable and accountable models.

Deeploy helps data scientists to deploy their models and makes them explainable, enabling interaction between humans and models. With Deeploy your ML deployments are manageable, accountable and explainable by design.