We present HUDD, a tool that supports safety analysis practices for systems enabled by Deep Neural Networks (DNNs) by automatically identifying the root causes for DNN errors and retraining the DNN. HUDD stands for Heatmap-based Unsupervised Debugging of DNNs, it automatically clusters error-inducing images whose results are due to common subsets of DNN neurons. The intent is for the generated clusters to group error-inducing images having common characteristics, that is, having a common root cause. HUDD identifies root causes by applying a clustering algorithm to matrices (i.e., heatmaps) capturing the relevance of every DNN neuron on the DNN outcome. Also, HUDD retrains DNNs with images that are automatically selected based on their relatedness to the identified image clusters. Our empirical evaluation with DNNs from the automotive domain have shown that HUDD automatically identifies all the distinct root causes of DNN errors, thus supporting safety analysis. Also, our retraining approach has shown to be more effective at improving DNN accuracy than existing approaches. A demo video of HUDD is available at https://youtu.be/drjVakP7jdU.

Heavy rain and blizzards aren't the only forms of severe weather Waymo's self-driving vehicles encounter on the regular. In a blog post published this morning, the Alphabet subsidiary laid out the ways its cars in over 25 cities tackle fog, dust, smoke, and other dangerous conditions that trip up even human drivers. "Challenging [environmental] conditions, which affect human driver and vehicle performance, are one of the leading contributors to crashes on our roads … Poor perception creates significant risk for other road users including pedestrians, cyclists, and other vehicle occupants," wrote Waymo chief safety officer Debbie Hersman. "Waymo is working hard to master a variety of weather scenarios as part of our mission to improve road safety." To this end, Waymo says its autonomous vehicles are designed to detect sudden extreme weather changes, such as a snowstorm, that could impact their ability to drive safely.