ALT Screenshot of paper. Title: Kepler: Robust Learning for Faster Parametric Query Optimization. Authors: LYRIC DOSHI∗, Google, USA VINCENT ZHUANG∗, Google, USA GAURAV JAIN, Google, USA RYAN MARCUS, University of Pennsylvania, USA HAOYU HUANG, Google, USA DENIZ ALTINBÜKEN, Google, USA EUGENE BREVDO, Google, USA CAMPBELL FRASER, Google, USA Abstract: Most existing parametric query optimization (PQO) techniques rely on traditional query optimizer cost models, which are often inaccurate and result in suboptimal query performance. We propose Kepler, an end-to-end learning-based approach to PQO that demonstrates significant speedups in query latency over a traditional query optimizer. Central to our method is Row Count Evolution (RCE), a novel plan generation algorithm based on perturbations in the sub-plan cardinality space. While previous approaches require accurate cost models, we bypass this requirement by evaluating candidate plans via actual execution data and training an ML model