LONDON, Feb. 18, 2020 /PRNewswire/ -- Mindtech Global Ltd, are pleased to announce that Chris Longstaff, VP Product Management, will present a talk "Addressing the Issues of Using 'Real-Data' for Training Visual AI Systems Through the Use of Synthetic Data" at the OpenTalks.AI conference, 20-21 February 2020, in Moscow.
OpenTalks.AI (https://opentalks.ai/en) is a leading conference on AI, bringing together anyone with an interest in AI, from scientists to entrepreneurs. A broad range of topics are discussed from the technical details through to the ethical and legal side of Artificial Intelligence.
During his talk, Chris will look at a number of the issues we have today in obtaining adequate, high quality, data for training neural networks. This talk will describe how synthetic data offers a solution to those shortcomings.
An example of the issues are the time and cost of obtaining data samples alongside the difficult issue of accurate and advanced annotations. The traditional method of gathering real-world data and annotating, cannot meet the requirements for training modern AI systems, whereas synthetic data is capable of generating vast amount of perfectly annotated data in a short time.
Ethical issues are another example of a key concern around today's Artificial Intelligence systems, and the training data can be just as important as during the deployment itself. Synthetic data helps these ethical issues in a number of ways: allowing training data to be created without risk of breach of privacy and the reduction of bias in the real data set by creating additional samples for under-represented groups in the data such as young children, or sufficient ethnic diversity.
Mindtech's Chameleon toolchain provides a full solution to the creation and management of synthetic data for training visual neural networks and is available for immediate licensing. The Chameleon tools are designed to help anyone implementing an AI based system to be able to generate the required training data in a fast, efficient manner, through its unique Scenario Editor that allows data scientists to immediately set up the required scenarios with a real-time preview of how the recorded data will appear. The Toolchain is being used today by partners in diverse applications including various Smart City and Smart Vision projects
For further information please contact firstname.lastname@example.org, or https://mindtech.global
SOURCE Mindtech Global Ltd