AvaWatz, a technology company that supports collaboration between robots, is proud to announce that it has just filed a patent application with the United States Patent and Trademark Office (USPTO). Titled “System and Method for Labeling, Evaluation, and Improvement of Training and Testing for Machine Learning,” the patent proposes a solution to the difficult problem of training data required for deep learning and machine learning models. The invention offers a cost-effective way to create high accuracy machine learning models – saving time and reducing the costs associated with training such systems.
Using AvaWatz’s approach, users can enhance their model’s performance by finding and correcting weaknesses in their training datasets, and improving the training of the model until it reaches desired results. This process has initially been used internally when dealing with challenging tasks such as testing the behaviour of robots amid extreme hazards or identifying tiny fragments of debris. However, AvaWatz is now offering their trusted AI services to other members of the Artificial Intelligence community.
AvaWatz is an AI firm specialising in creating adaptive robotics systems based on their Sense-Decide-Act paradigm, using advanced machine learning and algorithmic technologies. Dr. Rajini Anachi, the CEO of AvaWatz, underscores the importance of the company’s invention, noting that it should be particularly beneficial for those dealing with specialized use cases which require custom-made deep learning applications.
As well as caring for their customers, AvaWatz also values its team and will be proudly led by Dr. Rajini Anachi who is an AI expert with years of research and experience. She has made several significant contributions to the field, including projects with Intel, Honda, GE, and other major firms. She hopes that the patent application will encourage other developers in the intelligent automation space to create innovative solutions for efficient and cost-effective machine learning.
