Efficient Saliency Maps for Explainable AI
This makes fine-resolution saliency methods feasible for resource-limited platforms such as robots, cell phones, and industrial devices, addressing a bottleneck in deploying explainable AI in constrained environments.
The paper tackles the inefficiency of fine-resolution gradient methods for explainable AI saliency maps in deep CNNs by introducing a more efficient technique that measures information at each network scale and combines it into a single map, achieving similar or better accuracy with demonstrably superior results compared to methods like Guided Backprop and Grad-CAM++ without sacrificing speed.
We describe an explainable AI saliency map method for use with deep convolutional neural networks (CNN) that is much more efficient than popular fine-resolution gradient methods. It is also quantitatively similar or better in accuracy. Our technique works by measuring information at the end of each network scale which is then combined into a single saliency map. We describe how saliency measures can be made more efficient by exploiting Saliency Map Order Equivalence. We visualize individual scale/layer contributions by using a Layer Ordered Visualization of Information. This provides an interesting comparison of scale information contributions within the network not provided by other saliency map methods. Using our method instead of Guided Backprop, coarse-resolution class activation methods such as Grad-CAM and Grad-CAM++ seem to yield demonstrably superior results without sacrificing speed. This will make fine-resolution saliency methods feasible on resource limited platforms such as robots, cell phones, low-cost industrial devices, astronomy and satellite imagery.