MLCommons Consortium Chooses Cognata’s Synthetic Dataset to Set Industry Benchmarks

The MLCommons cognata’s benchmark dataset is now available for MLCommons members working on automotive benchmarks

Cognata LTD announces that MLCommons, a leading global engineering consortium, has officially selected Cognata’s high-precision AI simulation dataset for use in the MLPerf Automotive benchmark suite. Rooted in cutting-edge technology, Cognata’s automotive-grade, digital twin-based datasets will be used to measure industry benchmarks and improve practices in machine learning model training for the automotive industry.

Cognata’s datasets are based on a photorealistic simulation platform where virtual cars travel virtual roads, from cities to unmarked terrain, all true to real-world conditions. With advanced sensor modeling, Cognata’s AI-based Mobility traffic layers help engineers understand and analyze safety scenarios.

This collaboration fuses MLCommons’ commitment to improving machine learning for everyone with Cognata’s expertise in generating fully annotated, photorealistic synthetic datasets. These datasets are targeted for automotive sensors and are meticulously designed to detect vehicles and pedestrians. With over 20,000 unique frames, the dataset allows for full comparability and bias elimination.

Recommended AI News: Salary.com Acquires CompXL, The Enterprise Compensation Planning Leader

“We are excited to use the MLCommons Cognata dataset to define and develop our automotive ML benchmark suite”, said David Kanter, MLCommons Executive Director. “A high-quality dataset is an essential industry-standard benchmark that will help the entire industry move forward”.

Recommended AI News: Google Cloud and Spotify Expand Partnership to Help Unlock Creator Potential and Reach Fans

“Cognata’s digital twin based simulation platform for AI-based training, testing, and validation allows for infinite dataset generation, customized for the customer needs,” said Danny Atsmon, CEO and founder of Cognata. “Using this dataset to run benchmarks for improving ML and AI training with the highest fidelity is an important milestone in the automotive industry.”

[To share your insights with us, please write to sghosh@martechseries.com]