From the course: Natural Language Processing with ML.NET by Microsoft Press

Unlock this course with a free trial

Join today to access over 25,300 courses taught by industry experts.

Advanced concepts: Responsible AI

Advanced concepts: Responsible AI - ML.NET Tutorial

From the course: Natural Language Processing with ML.NET by Microsoft Press

Start my 1-month free trial Buy for my team

Advanced concepts: Responsible AI

“

- Leveraging tools like Azure Resources and GitHub facilities within our MLOps lifecycle is key. But when talking about AI applications, there's other pitfalls we should take into account as developers and which are intrinsic to the AI technology. Being aware of those risks and learn how to mitigate them is very important in production scenarios. In fact, AI may reinforce such biases without deliberate planning and design. For example, we might build a misogynistic model without deliberate planning, by training it on a biased dataset, representing mostly women cooking in a kitchen. As a result, the model will tend to recognize a woman also in a picture representing a man cooking. And this happens because machine learning models, during the data training process, try to extract patterns, in other words, correlations. For example, a model trained to recognize cars and motorcycles will find a correlation between the vehicle category and the number of wheels, easy, right? The problem is…

Contents