Ajay Thampi

About Me

Ajay Thampi, PhD in Signal Processing and Wireless Networks, is Lead Data Scientist at Microsoft.

Interpretable AI or How I Learned to Stop Worrying and Trust AI

By interpreting the model, we can gain a much deeper understanding and address problems like bias, leakage and trust. Interpretability is the degree to which a human can consistently estimate what a model will predict, how well the human can understand and follow the model’s prediction and finally, how well a human can detect when a model has made a mistake. It goes without saying that AI systems must be secure and safeguarded against adversarial attacks. 

The Harvard Innovation Lab

Made in Boston @

The Harvard Innovation Lab


Matching Providers

Matching providers 2
comments powered by Disqus.