This inventor holds 3 USPTO granted patents. Top assignees: Microsoft Technology Licensing, LLC, Other. Active years: 2014-2016.
Company Filing History:
2 out of 56,194 
1 out of 834,411 Years Active: 2014-2016
Title: Mikhail Kapralov: Innovator in Data Prediction and Graph Theory
Introduction
Mikhail Kapralov is a prominent inventor based in Stanford, CA (US). He has made significant contributions to the fields of data prediction and graph theory. With a total of 3 patents, his work focuses on enhancing the accuracy of predictions and optimizing graph structures.
Latest Patents
One of Kapralov's latest patents is titled "Generating a sparsifier using graph spanners." This invention involves creating a sparsifier from a union of multiple spanners of a graph. The edges of the sparsifier are weighted based on a measure of connectivity known as robust connectivity. The robust connectivity of a node pair is determined by the highest edge sampling probability at which the distance between the pair is likely to fall below a specified length. Each spanner is generated from a subgraph of the original graph using a decreasing sampling probability. The weight of each edge is set to the inverse of the estimated robust connectivity.
Another notable patent is "Predicting values in sequence." This invention utilizes multiple data prediction strategies to forecast the next data value in a sequence, each with a corresponding confidence value. Instead of relying on a single prediction strategy, Kapralov's method combines the predictions of each strategy to generate a more accurate overall prediction. Additionally, a deviation is calculated based on observed values in the sequence, favoring more recent values over less recent ones. This approach results in a prediction that is more precise and has a lower overall loss.
Career Highlights
Mikhail Kapralov has worked with notable companies, including Microsoft Technology Licensing, LLC. His experience in these organizations has contributed to his expertise in the fields of data prediction and graph theory.
Collaborations
One of his notable coworkers is Rina Panigrahy. Their collaboration has likely fostered innovative ideas and advancements in their respective fields.
Conclusion
Mikhail Kapralov is a distinguished inventor whose work in data prediction and graph theory has led to valuable patents. His innovative approaches continue to influence the landscape of technology and data analysis.