AI in Turbulent Times: A Webinar Series
We hope webinar series (and resources) will help organizations understand how to navigate the path forward during unprecedented times.
AI and machine learning models are trained on historical data, but COVID-19 has changed everything related to existing datasets. Companies are left with many questions, not the least of which is how to generate accurate models based on these extreme conditions. Listen to our AI in Turbulent Times where we break down how to approach AI during challenging times such as these.Watch the Webinar
Model Building: Using Small Datasets to Build Models
As the global coronavirus pandemic causes major disruptions to communities and the economy, many existing data science models struggle to adapt to these shifts due to a shortage of available data. Listen to our webinar, Using Small Datasets to Build Models, hosted by Customer-Facing Data Scientists Dave Heinicke and Rajiv Shah to learn:
- Strategies to build a “cold start” model
- Checks to ensure you have a meaningful, consistent signal from limited example
- Model insights that you can use to verify a meaningful model fit
New World, Old Model. Now What?
As the world changes and deployed machine learning models lose their power, you need to ensure that your predictions remain relevant and generate value. Join our Data Scientists, Peter Simon and Rajiv Shah, as they walk through the practical steps you can take in a world where the economy is shifting. We will cover:
- Assessing what has changed compared to when you trained your model
- Making quantitative assessments on how model drivers are changing
- Using economic history to inform your assumptions
- Tactics and strategies you can use to incorporate revised assumptions into your models
Managing Models in Uncertain Times
Monitoring and managing your production models in normal times is tricky enough. Listen to our webinar hosted by Rajiv Shah and Seph Mard to learn more about ML Ops. We will discuss proven and scalable methodologies for production model deployment, monitoring, and lifecycle management. We will address key questions such as:
- How do I maintain high-performing models in production?
- How do I know when my production models start to become unreliable?
- How do I quickly manage my models once their performance has decayed?
- How do I move fast while also increasing scrutiny, compliance, and transparency around your ML Ops processes?