Company Filing History:
Years Active: 2019-2021
Title: Innovations of Andreas Gerstlauer
Introduction
Andreas Gerstlauer is a prominent inventor based in Austin, TX (US). He has made significant contributions to the field of machine learning and graph processing. With a total of 2 patents, his work focuses on enhancing the efficiency and accuracy of predictive models.
Latest Patents
One of his latest patents is titled "Generating sets of training programs for machine learning models." This invention involves a method, system, and computer program product that generates training programs for machine learning models. It begins by receiving fixed values of workload metrics from a user, which correspond to low-level program features defining specific application behavior. A profile is created using these fixed values, and a suite of synthetic applications is generated to form a set of training programs targeting particular aspects of program behavior. This approach improves the prediction accuracy of machine learning-based predictive models by providing broader coverage of the program state-space.
Another notable patent is "Guided load balancing of graph processing workloads on heterogeneous clusters." This invention describes a method, system, and computer program product for load balancing graph processing workloads. It involves generating synthetic proxy graphs to characterize the graph processing speeds of machines in a cluster. Each graph application executing in the cluster is profiled using these synthetic graphs to form profiling sets. Metrics are computed from the relative speedup among the machines, and a natural graph is partitioned into multiple chunks based on selected metrics and user-defined partitioning algorithms.
Career Highlights
Andreas Gerstlauer is affiliated with the University of Texas System, where he continues to innovate and contribute to research in his field. His work has garnered attention for its practical applications in machine learning and graph processing.
Collaborations
He collaborates with notable colleagues such as Lizy Kurian John and Shuang Song, enhancing the research output and innovation potential within his team.
Conclusion
Andreas Gerstlauer's contributions to machine learning and graph processing through his patents reflect his commitment to advancing technology. His innovative approaches are paving the way for more efficient and accurate predictive models in various applications.