achine Learning & Advanced Analytics at Scale: A Platform Approach
Time: 10:40 am
day: Day One
- Data Science (DS) solutions need to lead to actions to realize value. Production DS is not only about algorithms: data pipeline development, feature engineering, model validation and lifecycle management are fundamental components.
- At Equinor, we are approaching DS from a platform perspective. We solve different problem sets with a common “mold”. As a result, we reduce our implementation costs. We call this methodology epsilon-cost scaling.
- This is being applied to operational planning and predictive maintenance for instance. Details will be discussed including:
- Data as a critical asset
- Platform approach to DS
- Epsilon-cost scaling advantages & constraints
- Focus on speed to build business impact