Popular Post


Monday, February 3, 2020

#XAI: my experience of arguing with #NeuralNet which I have built.

By reading the following paper I have learnt about XAI - eXplainable AI . Great idea, but very challenging.

Why asking an AI to explain itself can make things worse

That reading reminded me my personal interaction with AI. I  have shared my experience with building and applying Neural Network (NN) to recognize a different type of computer's workload, such as OLTP and others in my last year CMG Impact conference:

Catching Anomaly and Normality in Cloud by Neural Net and Entropy Calculation

I shared there the interesting observation that my NN model was not agreed with some of my labels in the training set:

What about you? Are you with me or with AI?

No comments:

Post a Comment