Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention

Jinkyu Kim, John Canny

Research output: Chapter in Book/Report/Conference proceedingConference contribution

162 Citations (Scopus)


Deep neural perception and control networks are likely to be a key component of self-driving vehicles. These models need to be explainable - they should provide easy-tointerpret rationales for their behavior - so that passengers, insurance companies, law enforcement, developers etc., can understand what triggered a particular behavior. Here we explore the use of visual explanations. These explanations take the form of real-time highlighted regions of an image that causally influence the network's output (steering control). Our approach is two-stage. In the first stage, we use a visual attention model to train a convolution network endto- end from images to steering angle. The attention model highlights image regions that potentially influence the network's output. Some of these are true influences, but some are spurious. We then apply a causal filtering step to determine which input regions actually influence the output. This produces more succinct visual explanations and more accurately exposes the network's behavior. We demonstrate the effectiveness of our model on three datasets totaling 16 hours of driving. We first show that training with attention does not degrade the performance of the end-to-end network. Then we show that the network causally cues on a variety of features that are used by humans while driving.

Original languageEnglish
Title of host publicationProceedings - 2017 IEEE International Conference on Computer Vision, ICCV 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages9
ISBN (Electronic)9781538610329
Publication statusPublished - 2017 Dec 22
Externally publishedYes
Event16th IEEE International Conference on Computer Vision, ICCV 2017 - Venice, Italy
Duration: 2017 Oct 222017 Oct 29

Publication series

NameProceedings of the IEEE International Conference on Computer Vision
ISSN (Print)1550-5499


Other16th IEEE International Conference on Computer Vision, ICCV 2017

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition


Dive into the research topics of 'Interpretable Learning for Self-Driving Cars by Visualizing Causal Attention'. Together they form a unique fingerprint.

Cite this