Diagnosis of Neural Network

Embodiments relate to systematic explanation of neural model behavior and effective deduction of its vulnerabilities. Input data is received for the neural model and applied to the model to generate output data. Accuracy of the output data is evaluated with respect to the neural model, and one or more neural model vulnerabilities are identified that correspond to the output data accuracy. An explanation of the output data and the identified one or more vulnerabilities is generated, wherein the explanation serves as an indicator of alignment of the input data with the output data.