AvaWatz Company (“AvaWatz”), a leading-edge AI company at the forefront of collaborative robotics, proudly announces that it has been awarded a contract by the U.S. Department of Homeland Security (DHS) for its Trusted AI product GENIE, codenamed “Clarifier”. This contract underscores the recognition of GENIE’s unparalleled capabilities in addressing fundamental challenges in machine learning (ML) and its potential to revolutionize critical sectors.
“We are honored to have been selected by the Department of Homeland Security for this important contract,” said Dr. Rajini Anachi, CEO of AvaWatz. “GENIE’s proven ability to drive down the operational overhead of achieving highly accurate ML models makes it an ideal solution for the agency’s mission-critical use cases requiring best-in-class AI performance.”
“GENIE provides a powerful suite of features that enhance efficiency, accuracy and reliability when training machine learning models. This helps organizations overcome the ‘long tail problem’ where achieving accuracy beyond 80% becomes exponentially costly and time-consuming.”
The ‘long tail problem’ refers to the challenge of diminishing returns in machine learning, where investment required to improve model accuracy becomes progressively greater as higher levels of accuracy are targeted. GENIE directly tackles this problem by significantly reducing the time and cost associated with training high-accuracy models through intelligent data labeling and dataset preparation capabilities.
GENIE enhances labeling efficiency through innovative features, including:
Initial Diverse Seed Selection: Laying a robust foundation by exposing models to a wide array of scenarios
Automatic Labeling: Significantly reducing manual effort via high-accuracy auto-labeling
Active Learning for Hard Samples: Focusing on the most informative data to learn complex patterns
Synthetic Data Generation and Augmentation: Enhancing dataset diversity for increased model robustness
Error Analysis on Unlabeled Data: Enabling continuous improvement without labeled test sets
Targeted Selection for Rare Classes/Slices: Addressing long-tail distributions for real-world accuracy
Filtering Outliers and Low-Quality Data: Ensuring models train only on reliable, high-quality data
The solution’s practical applications span multiple sectors where high-accuracy AI is critical, including healthcare for enhanced medical imaging, autonomous vehicles for detecting rare scenarios on the road, aviation for identifying runway debris, and energy for detecting insulator defects on power lines.
“Whether protecting national security, saving lives through medical imaging, or ensuring safe transportation, GENIE provides a powerful defensive solution against AI failure modes,” said Dr. Anachi. “We look forward to working with the Department of Homeland Security to enhance their AI initiatives.”
About AvaWatz
AvaWatz is an AI company specializing in advanced vision/perception, decision-making, and cooperative navigation for task execution using machine learning and physics-based algorithmic technologies. AvaWatz platform drives adaptive autonomous robotic systems built around a Detect-Decide-Act paradigm. AvaWatz provides End-to-End private/hybrid Cloud-based SaaS and real-time Edge AI solutions to public and private sector markets.
Learn more about AvaWatz investment opportunity here!