{"type":"ScholarlyArticle","title":"Concrete Problems for Autonomous Vehicle Safety: Advantages of Bayesian Deep Learning","authors":[{"name":"Rowan McAllister","slug":"rowan-mcallister","given":"Rowan","family":"McAllister","affiliation":"also Alan Turing Institute , London , UK"},{"name":"Yarin Gal","slug":"yarin-gal","given":"Yarin","family":"Gal","affiliation":"† Department of Engineering , University of Cambridge , UK"},{"name":"Alex Kendall","slug":"alex-kendall","given":"Alex","family":"Kendall","affiliation":"also Alan Turing Institute , London , UK"},{"name":"Mark van der Wilk","slug":"mark-van-der-wilk","given":"Mark","family":"van der Wilk","affiliation":"also Alan Turing Institute , London , UK"},{"name":"Amar Shah","slug":"amar-shah","given":"Amar","family":"Shah","affiliation":"also Alan Turing Institute , London , UK"},{"name":"Roberto Cipolla","slug":"roberto-cipolla","given":"Roberto","family":"Cipolla","affiliation":"also Alan Turing Institute , London , UK"},{"name":"Adrian Weller","slug":"adrian-weller","given":"Adrian","family":"Weller","affiliation":"† Department of Engineering , University of Cambridge , UK"}],"abstract":"Autonomous vehicle (AV) software is typically composed of a pipeline of individual components, linking sensor inputs to motor outputs. Erroneous component outputs propagate downstream, hence safe AV software must consider the ultimate effect of each component’s errors. Further, improving safety alone is not sufficient. Passengers must also feel safe to trust and use AV systems. To address such concerns, we investigate three under-explored themes for AV research: safety, interpretability, and compliance. Safety can be improved by quantifying the uncertainties of component outputs and propagating them forward through the pipeline. Interpretability is concerned with explaining what the AV observes and why it makes the decisions it does, building reassurance with the passenger. Compliance refers to maintaining some control for the passenger. We discuss open challenges for research within these themes. We highlight the need for concrete evaluation metrics, propose example problems, and highlight possible solutions.","doi":"10.24963/ijcai.2017/661","url":"https://ijcai.org/proceedings/2017/661","pdf":"https://ijcai.org/proceedings/2017/0661.pdf","markdown":"https://ijcai.org/proceedings/2017/661.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/661","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"AI and autonomy track","track":"AI and autonomy track","number":661,"pages":{"first":4745,"last":4753},"keywords":["Technical: Techniques","Technical: Models","Social: Challenges","Social: Human-machine Interaction"],"videos":[],"references":[{"position":1,"raw":"[Abbeel and Ng, 2004] Pieter Abbeel and Andrew Ng. Ap- prenticeship learning via inverse reinforcement learning. In ICML, page 1, 2004.","title":"Apprenticeship learning via inverse reinforcement learning","authors":"Ng Abbeel, Pieter Abbeel, Andrew Ng","year":"2004","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Aghasadeghi and Bretl, 2011] Navid Aghasadeghi and Timothy Bretl. Maximum entropy inverse reinforcement learning in continuous state spaces with path integrals. In IROS, pages 1561-1566, 2011.","title":"Maximum entropy inverse reinforcement learning in continuous state spaces with path integrals","authors":"Bretl ; ] Navid Aghasadeghi, Timothy Aghasadeghi, Bretl","year":"2011","venue":"IROS","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Amodei et al., 2016] Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, and Dan Mané. Concrete problems in AI safety. arXiv preprint arXiv:1606.06565, 2016.","title":"Concrete problems in AI safety","authors":"Dario Amodei, Chris Olah, Jacob Steinhardt, Paul Christiano, John Schulman, Dan Mané","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Astrom, 1965] Karl Astrom. Optimal control of Markov de- cision processes with incomplete state estimation. Journal of Mathematical Analysis and Applications, 10(1):174- 205, 1965.","title":"Optimal control of Markov decision processes with incomplete state estimation","authors":"Karl Astrom, Astrom","year":"1965","venue":"Journal of Mathematical Analysis and Applications","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Bandyopadhyay et al., 2013] Tirthankar Bandyopadhyay, Kok Won, Emilio Frazzoli, David Hsu, Wee Lee, and Daniela Rus. Intention-Aware Motion Planning, pages 475-491. Berlin, Heidelberg, 2013.","title":"Intention-Aware Motion Planning","authors":"Tirthankar Bandyopadhyay, Emilio Kok Won, David Frazzoli, Wee Hsu, Daniela Lee, Rus","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Baronas and Louis, 1988] Ann-Marie Baronas and Meryl Louis. Restoring a sense of control during implementa- tion: How user involvement leads to system acceptance. MIS Quarterly, 12(1):111-124, 1988.","title":"Restoring a sense of control during implementation: How user involvement leads to system acceptance","authors":"Ann-Marie Baronas, Meryl Louis","year":"1988","venue":"MIS Quarterly","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Bast et al., 2016] Hannah Bast, Daniel Delling, Andrew Goldberg, Matthias Müller-Hannemann, Thomas Pajor, Peter Sanders, Dorothea Wagner, and Renato Werneck. Route planning in transportation networks. In Algorithm Engineering, pages 19-80. 2016.","title":"Route planning in transportation networks","authors":"Hannah Bast, Daniel Delling, Andrew Goldberg, Matthias Müller-Hannemann, Thomas Pajor, Peter Sanders, Dorothea Wagner, Renato Werneck","year":"2016","venue":"Algorithm Engineering","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Berger, 2013] James Berger. Statistical decision theory and Bayesian analysis. Springer Science & Business Media, 2013.","title":"Statistical decision theory and Bayesian analysis","authors":"James Berger","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Bernardo and Smith, 2001] José M Bernardo and Adrian Smith. Bayesian theory. Measurement Science and Tech- nology, 12(2):221, 2001.","title":"José M Bernardo and Adrian Smith","authors":"Smith Bernardo","year":"2001","venue":"Measurement Science and Technology","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Bojarski et al., 2017] Mariusz Bojarski, Philip Yeres, Anna Choromanska, Krzysztof Choromanski, Bernhard Firner, Lawrence Jackel, and Urs Muller. Explaining how a deep neural network trained with end-to-end learning steers a car. arXiv preprint arXiv:1704.07911, 2017.","title":"Explaining how a deep neural network trained with end-to-end learning steers a car","authors":"Mariusz Bojarski, Philip Yeres, Anna Choromanska, Krzysztof Choromanski, Bernhard Firner, Lawrence Jackel, Urs Muller","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Brunner et al., 2013] Christopher Brunner, Thierry Peynot, Teresa Vidal-Calleja, and James Underwood. Selective combination of visual and thermal imaging for resilient localization in adverse conditions: Day and night, smoke and fire. Journal of Field Robotics, 30(4):641-666, 2013.","title":"Selective combination of visual and thermal imaging for resilient localization in adverse conditions: Day and night, smoke and fire","authors":"Christopher Brunner, Thierry Peynot, Teresa Vidal-Calleja, James Underwood","year":"2013","venue":"Journal of Field Robotics","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Bry and Roy, 2011] Adam Bry and Nicholas Roy. Rapidly- exploring random belief trees for motion planning under uncertainty. In ICRA, pages 723-730, May 2011.","title":"Rapidlyexploring random belief trees for motion planning under uncertainty","authors":"Roy ; Bry, Adam Bry, Nicholas Roy","year":"2011","venue":"ICRA","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Dabkowski and Gal, 2017] Piotr Dabkowski and Yarin Gal. Real time image saliency for black box classifiers. 2017.","title":"Real time image saliency for black box classifiers","authors":"Piotr Dabkowski, Yarin Gal","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Deisenroth and Rasmussen, 2011] Marc Deisenroth and Carl Rasmussen. Pilco: A model-based and data-efficient approach to policy search. In ICML, pages 465-472, 2011.","title":"Pilco: A model-based and data-efficient approach to policy search","authors":"Rasmussen Deisenroth, Marc Deisenroth, Carl Rasmussen","year":"2011","venue":"ICML","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Depeweg et al., 2016] Stefan Depeweg, José Miguel Hernández-Lobato, Finale Doshi-Velez, and Steffen Ud- luft. Learning and policy search in stochastic dynamical systems with Bayesian neural networks. arXiv preprint arXiv:1605.07127, 2016.","title":"Learning and policy search in stochastic dynamical systems with Bayesian neural networks","authors":"Stefan Depeweg, José Miguel Hernández-Lobato, Finale Doshi-Velez, Steffen Udluft","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Doshi-Velez and Kim, 2017] Finale Doshi-Velez and Been Kim. A roadmap for a rigorous science of interpretability. arXiv preprint arXiv:1702.08608, 2017.","title":"Finale Doshi-Velez and Been Kim. A roadmap for a rigorous science of interpretability","authors":"Kim Doshi-Velez","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Durrant-Whyte and Bailey, 2006] Hugh Durrant-Whyte and Tim Bailey. Simultaneous localization and map- ping: part i. IEEE robotics & automation magazine, 13(2):99-110, 2006.","title":"Simultaneous localization and mapping: part i","authors":"Durrant-Whyte, Bailey, Hugh Durrant-Whyte, Tim Bailey","year":"2006","venue":"IEEE robotics & automation magazine","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Durrant-Whyte, 1988] Hugh Durrant-Whyte. Uncertain ge- ometry in robotics. Journal on Robotics and Automation, 4(1):23-31, 1988.","title":"Uncertain geometry in robotics","authors":"Durrant-Whyte ;, Hugh Durrant-Whyte","year":"1988","venue":"Journal on Robotics and Automation","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Eigen and Fergus, 2015] David Eigen and Rob Fergus. Pre- dicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture. In ICCV, pages 2650-2658, 2015.","title":"Predicting depth, surface normals and semantic labels with a common multi-scale convolutional architecture","authors":"David Eigen, Rob Fergus","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Gal et al., 2016] Yarin Gal, Rowan McAllister, and Carl Rasmussen. Improving PILCO with Bayesian neural net- work dynamics models. In Data-Efficient Machine Learn- ing workshop, ICML, 2016.","title":"Improving PILCO with Bayesian neural network dynamics models","authors":"Yarin Gal, Rowan Mcallister, Carl Rasmussen","year":"2016","venue":"Data-Efficient Machine Learning workshop","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Gal, 2016] Yarin Gal. Uncertainty in Deep Learning. PhD thesis, University of Cambridge, 2016.","title":"Uncertainty in Deep Learning","authors":"Yarin Gal","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Garg et al., 2016] Ravi Garg, Gustavo Carneiro, and Ian Reid. Unsupervised cnn for single view depth estimation: Geometry to the rescue. In European Conference on Com- puter Vision, pages 740-756. Springer, 2016.","title":"Unsupervised cnn for single view depth estimation: Geometry to the rescue","authors":"Ravi Garg, Gustavo Carneiro, Ian Reid","year":"2016","venue":"European Conference on Computer Vision","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Giffi and Vitale, 2017] Craig Giffi and Joe Vitale. Fact, fic- tion and fear cloud future of autonomous vehicles in the minds of consumers. Technical report, Deloitte, Jan 2017.","title":"Fact, fiction and fear cloud future of autonomous vehicles in the minds of consumers","authors":"Craig Giffi, Joe Vitale","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Isard and Blake, 1998] Michael Isard and Andrew Blake. Condensation-conditional density propagation for visual tracking. IJCV, 29(1):5-28, 1998.","title":"Condensation-conditional density propagation for visual tracking","authors":"Michael Isard, Andrew Blake","year":"1998","venue":"IJCV","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Julier et al., 1995] Simon Julier, Jeffrey Uhlmann, and Hugh Durrant-Whyte. A new approach for filtering nonlin- ear systems. In American Control Conference, volume 3, pages 1628-1632, Jun 1995.","title":"A new approach for filtering nonlinear systems","authors":"Simon Julier, Jeffrey Uhlmann, Hugh Durrant-Whyte","year":"1995","venue":"American Control Conference","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Kalman, 1960] Rudolph Kalman. A new approach to linear filtering and prediction problems. Journal of Basic Engi- neering, 82(1):35-45, 1960.","title":"A new approach to linear filtering and prediction problems","authors":"Rudolph Kalman","year":"1960","venue":"Journal of Basic Engineering","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Kendall and Cipolla, 2016] Alex Kendall and Roberto Cipolla. Modelling uncertainty in deep learning for camera relocalization. ICRA, 2016.","title":"Modelling uncertainty in deep learning for camera relocalization","authors":"Alex Kendall, Roberto Cipolla","year":"2016","venue":"ICRA","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Kendall and Gal, 2017] Alex Kendall and Yarin Gal. What uncertainties do we need in Bayesian deep learning for computer vision? arXiv preprint arXiv:1703.04977, 2017.","title":"What uncertainties do we need in Bayesian deep learning for computer vision","authors":"Alex Kendall, Yarin Gal","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Kendall et al., 2015] Alex Kendall, Vijay Badrinarayanan, and Roberto Cipolla. Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding. arXiv preprint arXiv:1511.02680, 2015.","title":"Bayesian segnet: Model uncertainty in deep convolutional encoder-decoder architectures for scene understanding","authors":"Alex Kendall, Vijay Badrinarayanan, Roberto Cipolla","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Kendall et al., 2017] Alex Kendall, Yarin Gal, and Roberto Cipolla. Multi-task learning using uncertainty to weigh losses for scene geometry and semantics. 2017.","title":"Multi-task learning using uncertainty to weigh losses for scene geometry and semantics","authors":"Alex Kendall, Yarin Gal, Roberto Cipolla","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":31,"raw":"[LaValle, 2006] Steven LaValle. Planning algorithms. Cam- bridge University Press, 2006.","title":"Planning algorithms","authors":"Steven Lavalle","year":"2006","venue":null,"doi":null,"url":null,"internal_url":null},{"position":32,"raw":"[Levine et al., 2010] Sergey Levine, Zoran Popovic, and Vladlen Koltun. Feature construction for inverse reinforce- ment learning. In NIPS, pages 1342-1350, 2010.","title":"Feature construction for inverse reinforcement learning","authors":"Sergey Levine, Zoran Popovic, Vladlen Koltun","year":"2010","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":33,"raw":"[Levine et al., 2011] Sergey Levine, Zoran Popovic, and Vladlen Koltun. Nonlinear inverse reinforcement learning with Gaussian processes. In NIPS, pages 19-27, 2011.","title":"Nonlinear inverse reinforcement learning with Gaussian processes","authors":"Sergey Levine, Zoran Popovic, Vladlen Koltun","year":"2011","venue":"NIPS","doi":null,"url":null,"internal_url":null},{"position":34,"raw":"[Luettel et al., 2012] Thorsten Luettel, Michael Himmels- bach, and Hans-Joachim Wuensche. Autonomous ground vehicles -concepts and a path to the future. Proceedings of the IEEE, 100:1831-1839, May 2012.","title":"Autonomous ground vehicles -concepts and a path to the future","authors":"Luettel","year":"2012","venue":"Proceedings of the IEEE","doi":null,"url":null,"internal_url":null},{"position":35,"raw":"[McAllister et al., 2012] Rowan McAllister, Thierry Peynot, Robert Fitch, and Salah Sukkarieh. Motion planning and stochastic control with experimental validation on a plan- etary rover. In IROS, pages 4716-4723, Oct 2012.","title":"Motion planning and stochastic control with experimental validation on a planetary rover","authors":"Rowan Mcallister, Thierry Peynot, Robert Fitch, Salah Sukkarieh","year":"2012","venue":"IROS","doi":null,"url":null,"internal_url":null},{"position":36,"raw":"[Melchior and Simmons, 2007] Nik Melchior and Reid Sim- mons. Particle RRT for path planning with uncertainty. In ICRA, pages 1617-1624. IEEE, 2007.","title":"Particle RRT for path planning with uncertainty","authors":"Nik Melchior, Reid Simmons","year":"2007","venue":"ICRA","doi":null,"url":null,"internal_url":null},{"position":37,"raw":"[Neverova et al., 2017] Natalia Neverova, Pauline Luc, Camille Couprie, Jakob Verbeek, and Yann LeCun. Pre- dicting deeper into the future of semantic segmentation. arXiv preprint arXiv:1703.07684, 2017.","title":"Predicting deeper into the future of semantic segmentation","authors":"Natalia Neverova, Pauline Luc, Camille Couprie, Jakob Verbeek, Yann Lecun","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":38,"raw":"[Ng and Russell, 2000] Andrew Ng and Stuart Russell. Al- gorithms for inverse reinforcement learning. In ICML, pages 663-670, 2000. [NHTSA, 2008] NHTSA. Dot hs 811 059: National mo- tor vehicle crash causation survey. Technical report, U.S. Department of Transportation, National Highway Traffic Safety Administration, Jul 2008. [NHTSA, 2017] NHTSA. PE 16-007. Technical report, U.S. Department of Transportation, National Highway Traffic Safety Administration, Jan 2017. Tesla Crash Preliminary Evaluation Report. [NSC, 2016] NSC. Motor vehicle fatality estimates. Techni- cal report, National Safety Council, Statistics Department, Feb 2016.","title":"Algorithms for inverse reinforcement learning","authors":"Russell ; Ng, Andrew Ng, Stuart Russell","year":"2000","venue":"National motor vehicle crash causation survey","doi":null,"url":null,"internal_url":null},{"position":39,"raw":"[Paden et al., 2016] Brian Paden, Michal Čáp, Sze Zheng Yong, Dmitry Yershov, and Emilio Frazzoli. A sur- vey of motion planning and control techniques for self- driving urban vehicles. Transactions on Intelligent Vehi- cles, 1(1):33-55, 2016.","title":"A survey of motion planning and control techniques for selfdriving urban vehicles","authors":"Brian Paden, Michal Čáp, Zheng Sze, Dmitry Yong, Emilio Yershov, Frazzoli","year":"2016","venue":"Transactions on Intelligent Vehicles","doi":null,"url":null,"internal_url":null},{"position":40,"raw":"[Ramachandran and Amir, 2007] Deepak Ramachandran and Eyal Amir. Bayesian inverse reinforcement learning. Urbana, 51(61801):1-4, 2007.","title":"Bayesian inverse reinforcement learning","authors":"Deepak Ramachandran, Eyal Amir","year":"2007","venue":"Urbana","doi":null,"url":null,"internal_url":"/proceedings/2007/416"},{"position":41,"raw":"[Ribeiro et al., 2016] Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. Why should i trust you?: Explain- ing the predictions of any classifier. In KDD, pages 1135- 1144, 2016.","title":"Why should i trust you?: Explaining the predictions of any classifier","authors":"Marco Tulio Ribeiro, Sameer Singh, Carlos Guestrin","year":"2016","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":42,"raw":"[SAE, 2014] SAE. J3016 surfact vehicle information report. Technical report, Society of Automotive Engineers Inter- national, Jan 2014. Levels of driving automation.","title":"SAE. J3016 surfact vehicle information report","authors":null,"year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":43,"raw":"[Simonyan et al., 2013] Karen Simonyan, Andrea Vedaldi, and Andrew Zisserman. Deep inside convolutional networks: Visualising image classification models and saliency maps. arXiv preprint arXiv:1312.6034, 2013.","title":"Deep inside convolutional networks: Visualising image classification models and saliency maps","authors":"Karen Simonyan, Andrea Vedaldi, Andrew Zisserman","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":44,"raw":"[Sivaraman and Trivedi, 2013] Sayanan Sivaraman and Mo- han Manubhai Trivedi. Looking at vehicles on the road: A survey of vision-based vehicle detection, tracking, and behavior analysis. Transactions on Intelligent Transporta- tion Systems, 14(4):1773-1795, Dec 2013.","title":"Looking at vehicles on the road: A survey of vision-based vehicle detection, tracking, and behavior analysis","authors":"Trivedi Sivaraman, Sayanan Sivaraman, Mohan Manubhai Trivedi","year":"2013","venue":"Transactions on Intelligent Transportation Systems","doi":null,"url":null,"internal_url":null},{"position":45,"raw":"[Sondik, 1971] Edward Sondik. The optimal control of par- tially observable Markov processes. Technical report, 1971.","title":"The optimal control of partially observable Markov processes","authors":"Edward Sondik","year":"1971","venue":null,"doi":null,"url":null,"internal_url":null},{"position":46,"raw":"[Štrumbelj and Kononenko, 2014] Erik Štrumbelj and Igor Kononenko. Explaining prediction models and individ- ual predictions with feature contributions. Knowledge and information systems, 41(3):647-665, 2014.","title":"Erik Štrumbelj and Igor Kononenko. Explaining prediction models and individual predictions with feature contributions","authors":"Kononenko Štrumbelj","year":"2014","venue":"Knowledge and information systems","doi":null,"url":null,"internal_url":null},{"position":47,"raw":"[Thrun et al., 2005] Sebastian Thrun, Wolfram Burgard, and Dieter Fox. Probabilistic robotics. MIT press, 2005.","title":"Probabilistic robotics","authors":"Sebastian Thrun, Wolfram Burgard, Dieter Fox","year":"2005","venue":null,"doi":null,"url":null,"internal_url":null},{"position":48,"raw":"[WHO, 2015] WHO. Global status report on road safety 2015. World Health Organization, Violence and Injury Prevention, 2015.","title":"WHO. Global status report on road safety 2015","authors":null,"year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":49,"raw":"[Wulfmeier et al., 2015a] Markus Wulfmeier, Peter On- druska, and Ingmar Posner. Deep inverse reinforcement learning. CoRR, 2015.","title":"Deep inverse reinforcement learning","authors":"Wulfmeier","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":50,"raw":"[Wulfmeier et al., 2015b] Markus Wulfmeier, Peter On- druska, and Ingmar Posner. Maximum entropy deep inverse reinforcement learning. arXiv preprint arXiv:1507.04888, 2015.","title":"Maximum entropy deep inverse reinforcement learning","authors":"Wulfmeier","year":"2015","venue":null,"doi":null,"url":null,"internal_url":null},{"position":51,"raw":"[Zeiler and Fergus, 2014] Matthew Zeiler and Rob Fergus. Visualizing and understanding convolutional networks. In European conference on computer vision, pages 818-833, 2014.","title":"Visualizing and understanding convolutional networks","authors":"Matthew Zeiler, Rob Fergus","year":"2014","venue":"European conference on computer vision","doi":null,"url":null,"internal_url":null},{"position":52,"raw":"[Zhou et al., 2014] Bolei Zhou, Aditya Khosla, Agata Lapedriza, Aude Oliva, and Antonio Torralba. Object detectors emerge in deep scene CNNs. arXiv preprint arXiv:1412.6856, 2014.","title":"Object detectors emerge in deep scene CNNs","authors":"Zhou","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":53,"raw":"[Ziebart et al., 2008] Brian Ziebart, Andrew Maas, Andrew Bagnell, and Anind Dey. Maximum entropy inverse rein- forcement learning. In AAAI, volume 8, pages 1433-1438, 2008.","title":"Maximum entropy inverse reinforcement learning","authors":"Brian Ziebart, Andrew Maas, Andrew Bagnell, Anind Dey","year":"2008","venue":"AAAI","doi":null,"url":null,"internal_url":null},{"position":54,"raw":"[Ziebart et al., 2009] Brian Ziebart, Nathan Ratliff, Garratt Gallagher, Christoph Mertz, Kevin Peterson, Andrew Bag- nell, Martial Hebert, Anind Dey, and Siddhartha Srinivasa. Planning-based prediction for pedestrians. In IROS, pages 3931-3936, 2009.","title":"Planning-based prediction for pedestrians","authors":"Brian Ziebart, Nathan Ratliff, Garratt Gallagher, Christoph Mertz, Kevin Peterson, Andrew Bagnell, Martial Hebert, Anind Dey, Siddhartha Srinivasa","year":"2009","venue":"IROS","doi":null,"url":null,"internal_url":null},{"position":55,"raw":"[Zintgraf et al., 2017] Luisa Zintgraf, Taco Cohen, Tameem Adel, and Max Welling. Visualizing deep neural network decisions: Prediction difference analysis. arXiv preprint arXiv:1702.04595, 2017.","title":"Visualizing deep neural network decisions: Prediction difference analysis","authors":"Luisa Zintgraf, Taco Cohen, Tameem Adel, Max Welling","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}