Abner Guzman-Rivera

Papers

Total Citations

H-Index

About

Abner Guzman-Rivera is a researcher whose work sits at the intersection of probabilistic machine learning and relational reasoning. His most cited contribution, "Lifted Relational Kalman Filtering" (2013, 26 citations), addresses a fundamental challenge in dynamic systems: scaling Kalman Filters—a cornerstone tool in robotics, forecasting, and defense—to complex, relational domains. By introducing lifted inference techniques, Guzman-Rivera enabled efficient state estimation when variables share symmetries, significantly reducing computational overhead compared to traditional approaches. This work bridges the gap between statistical relational learning and filtering, offering practical advances for applications requiring real-time reasoning over interconnected variables. Beyond this, his research explores structured prediction and learning under uncertainty, with contributions that emphasize both theoretical rigor and algorithmic efficiency. While his citation count reflects a focused, impactful body of work, the conceptual depth of his contributions—particularly in making probabilistic inference tractable for relational systems—positions him as a thoughtful innovator in machine learning. For students and researchers, Guzman-Rivera’s work exemplifies how blending classical tools with modern relational frameworks can unlock new capabilities in dynamic, data-rich environments.