Shima Imani

Title: Shima Imani: Innovator in Mathematical Reasoning and Time Series Segmentation

Introduction

Shima Imani is a notable inventor based in Sammamish, WA (US). He has made significant contributions to the fields of artificial intelligence and data analysis, holding 2 patents that showcase his innovative approach to solving complex problems.

Latest Patents

One of Shima Imani's latest patents is titled "Mathematical reasoning using large language models." This patent discloses techniques for an AI system utilizing a large language model (LLM) to improve accuracy and reliability in solving mathematical problems. The process involves transforming an initial query into a template query by replacing original input values with variables. Multiple prompts are sent to the LLM, each differing and contextually related to the template query. The system receives multiple results from the LLM, each containing an analytical expression to solve the mathematical problem. These expressions are evaluated using a numerical evaluation tool, and a consensus is achieved when the evaluated expressions satisfy a consensus condition. Once this condition is met, the original inputs are evaluated with one or more of the expressions, leading to the final solution.

Another significant patent by Shima Imani is "Univariate time series segmentation using proxy variables and sparse graph recovery algorithms." This patent relates to a time series segmentation system that efficiently segments univariate time series data. The system utilizes proxy variable time series to identify distinct segments within a univariate time series. It generates proxy variables that approximate the univariate time series and combines them to create a supplemented multivariate time series. The system then divides this supplemented multivariate time series into portions using time-based windows, converts the windowed subsequences into graph objects, and utilizes a conditional similarity model to determine segmentation timestamps. Ultimately, a segmented univariate time series is generated from these timestamps.

Career Highlights

Shima Imani is currently employed at Microsoft Technology Licensing, LLC, where he continues to innovate and contribute to advancements in technology. His work focuses on enhancing AI systems and improving data analysis techniques.

Collaborations

Shima has collaborated with talented individuals such as Harsh Shrivastava and Liang Du, further enriching his work and expanding the impact of his inventions.

Conclusion

Shima Imani is a distinguished inventor whose work in mathematical reasoning and time series segmentation has the potential to revolutionize the way we