Introducing Metric Management

Date: Thursday, May 28

Time: 10am PT / 1pm ET

SigOpt provides a “human-in-the-loop” process so that you can guide the platform’s discovery of a more effective set of parameters for your model, based on external business factors or domain-specific expertise. This process is called Metric Management, and is enabled by a variety of tools built into our API and web interface.

In this webinar, we will introduce Metric Management, which includes the following advanced features:

  • Stored Metrics: store up to 100 metrics for every training run, for later analysis
  • Metric Thresholds: establish minimums or maximums for what counts as a valid objective
  • Metric Strategy: select your goal via API, as to how you want to assess your metrics
  • Metric Failures: automatically report failures to SigOpt to help guide the optimization process away from regions that don’t produce a converged model
  • Multimetric Optimization: select up to two metrics to optimize against, and then optionally optimize against a new second metric
  • Metric Constraints (newly generally available): provide a bounding function around where you want SigOpt to search, and update it while your model is tuned

We will give an overview and a live demo of each feature, and show how they are more valuable when used in concert rather than independently. Sign up for the webinar to learn more!


Barrett Williams

Barrett crafts product messaging for both technical practitioners and senior decision makers. He is a graduate of Yale University where he studied economics and international studies. He also studied electrical engineering at Columbia, prototyped augmented reality glasses at Meta, marketed robotics developer kits at NVIDIA, and wrote launch and evergreen content for machine learning products at Google Cloud.



Dr. Harvey Cheng

Harvey is a research engineer at SigOpt with a strong interest in stochastic optimization and machine learning. At SigOpt, he applies these interests to develop novel implementations of optimization algorithms for enterprise and academic users. Prior to SigOpt, Harvey obtained his Ph.D. in electrical engineering from Princeton University, where his doctoral studies focused on approximate dynamic programming, stochastic optimization, and optimal learning with applications in managing grid-level battery storage. He also holds a B.S. in electrical engineering from the University of Texas at Austin.