New Releases, Enhancements, Changes + Arize in the News!


Model Performance Tab - Click to Add/Remove Cohort as a Filter

Select a feature value in the performance breakdown distribution to add or remove that cohort as a filter. This enables you to quickly debug whether removing a particular slice of data improves performance or if you need to dig deeper into that particular slice (feature-value combination) of data.

Drift Monitors: Top K

You can now monitor the drift of only the top K percent of your feature values and see how the distribution has changed.

In the News

Best Practices In ML Observability for Customer Lifetime Value (LTV) Models

Lifetime Value is essential to evaluate the total worth of a customer to a business over their whole relationship, and using ML observability can help ensure the predictions of LTV models stay in tip-top shape. Read More.

How ML Observability Helps America First Credit Union Stay a Step Ahead

Last month, we hosted a Q&A webinar with our client, America First Credit Union. Their data science lead, Richard Woolston shared with the audience how they are able to leverage ML observability to stay ahead in a competitive market. If you weren't able to attend, no problem! Check out our recap and takeaways. Read More.

Introducing Remi Cattiau, Arize’s Chief Information Security Officer

We are so excited to introduce Remi Cattiau, Arize AI’s Chief Information Security Officer (CISO). Joining Arize in November, Remi brings nearly two decades of experience as a developer, consultant, and executive charged overseeing cloud security for large enterprise clients. Learn more about his career journey and his role at Arize. Read More.

What Are the Prevailing Explainability Methods?

As models increase in complexity, the ability to introspect and understand why a model made a particular prediction becomes more and more difficult. So what are today's prevailing explainability methods you ask? Learn More.

Last updated