We propose a new attribution method for neural networks developed using first principles of causality (to the best of our knowledge, the first such). The neural network architecture is viewed as a Structural Causal Model, and a methodology to compute the causal effect of each feature on the output is presented. With reasonable assumptions on the causal structure of the input data,we propose algorithms to efficiently compute the causal effects, as well as scale the approach to data with large dimensionality. We also show how this method can be used for recurrent neural networks.We report experimental results on both simulated and real datasets showcasing the promise and usefulness of the proposed algorithm. Read More
Daily Archives: September 10, 2019
World Economic Forum – 8 predictions on AI & Robotics
The Rise of T-1000: Artificial Intelligence on the Battlefield
Artificial Intelligence (AI) or machine learning is being used by military intelligence and at the high strategic level. The question is whether this technology will ever filter down to the soldier actually doing the fighting on the ground? Science fiction novels and movies suggest a system that can communicate with a warfighter in real time and provide situational awareness, but how far is the fiction from reality?
“The most important weapon is situation awareness, and there are AI-based tools to help a lot with this,” explained Jim Purtilo, associate professor of computer science at the University of Maryland. Read More
Rick's Cafe AI 