Search Engine technology giant launches TensorFlow Constrained Optimization (TFCO), which is a supervised machine learning library for training the machine learning models. This library is used for optimizing inequality-constrained problems in TensorFlow 1.14 and in TensorFlow 2.
The TensorFlow Constrained Optimization (TFCO) library is used for training machine learning model using multiple metrics and “optimizing inequality-constrained problems.”
This machine learning library is designed to help the data scientist in addressing the fairness constrains and predictive parity. This library will help in solving the issues related to the fairness of the model while training machine learning models.
This library was tested with the Wikipedia data set and the library was able to lower false-positive rates while predicting the predicting the user comments toxicity based on race, religion, gender identity, or sexuality, while maintaining similar accuracy rates.
TFCO is made to “take into account the societal and cultural factors necessary to satisfy real-world requirements,” said Andrew Zaldivar on behalf of the TFCO team today in a Google AI blog post.
“The ability to express many fairness goals as rate constraints can help drive progress in the responsible development of machine learning, but it also requires developers to carefully consider the problem they are trying to address,” he said. “A ‘safer’ alternative is to constrain each group to independently satisfy some absolute metric, for example by requiring each group to achieve at least 75% accuracy. Using such absolute constraints rather than relative constraints will generally keep the groups from dragging each other down.”
This also comes with the optional “two-dataset” approach to improving generalization, which is built on a trio of research paper publish in 2019.
This library is available at github and can be accessed by visiting at: TensorFlow Constrained Optimization (TFCO).
Ads: Do you want to learn latest version of TensorFlow and master the machine learn with example codes? Check the tutorials at TensorFlow 2.0 Tutorials and Example. TensorFlow 2.1.0 is release last mont , check TensorFlow 2.1.0 features.
