IEEE W Michigan CS/RA Technical talk-AI Explainability Questions-Justifying ML Predictions : vTools eNotice

IEEE.org | IEEE Xplore Digital Library | IEEE Standards | IEEE Spectrum | More Sites

IEEE W Michigan CS/RA Technical talk-AI Explainability Questions-Justifying ML Predictions

To view complete details for this event, click here to view the announcement

IEEE West Michigan CS/RA Technical talk: AI Explainability Challenges: Justifying ML Model Predictions

The IEEE West Michigan joint Computer and Robotics & Automation society (CS/RA) would like to invite you to attend our technical webinar titled “Challenges with Artificial intelligence Explainability: Justifying Machine learning model predictions”.

The webinar is presented by Dr. Mohamed Kalil from IBM Watson Analytics.

Abstract

Machine learning models are all around us these days. They help us predict weather based on historic data, they help our employers predict business metrics, and they help our phones predict the word we will type next. While ML models have achieved great success and a high degree of accuracy in many areas, they can still be surrounded by mystery. Some models can be easily interpreted, and their predictions explained. But many models still feel like a black box. They share their output but not much more in terms of the detailed reasoning.

This can have a high impact, especially with the roles of AI models getting more significant. The models are moving from supporting convenience features, to matters that impact Human life like self-driving cars or the law like detecting and analyzing crime scenes.

This webinar will discuss the significance of various ML/AI models and the latest on Artificial intelligence explainability (AIX).

Speaker

Dr. Kalil is a data scientist and an AI researcher at IBM Watson analytics. He has distinguished experience in the field of Artificial Intelligence as seen in various publications apanning the topics in the areas of optimization, business data analytics and network performance tuning. His recent interests includes large scale business analytics models and Artificial Intelligence model explainability.