{"type":"ScholarlyArticle","title":"Adversarial Generation of Real-time Feedback with Neural Networks for Simulation-based Training","authors":[{"name":"Xingjun Ma","slug":"xingjun-ma","given":"Xingjun","family":"Ma","affiliation":"The University of Melbourne , Australia"},{"name":"Sudanthi Wijewickrema","slug":"sudanthi-wijewickrema","given":"Sudanthi","family":"Wijewickrema","affiliation":"The University of Melbourne , Australia"},{"name":"Shuo Zhou","slug":"shuo-zhou","given":"Shuo","family":"Zhou","affiliation":"The University of Melbourne , Australia"},{"name":"Yun Zhou","slug":"yun-zhou","given":"Yun","family":"Zhou","affiliation":"The University of Melbourne , Australia"},{"name":"Zakaria Mhammedi","slug":"zakaria-mhammedi","given":"Zakaria","family":"Mhammedi","affiliation":"The University of Melbourne , Australia"},{"name":"Stephen O'Leary","slug":"stephen-o-leary","given":"Stephen","family":"O'Leary","affiliation":"The University of Melbourne , Australia"},{"name":"James Bailey","slug":"james-bailey","given":"James","family":"Bailey","affiliation":"The University of Melbourne , Australia"}],"abstract":"Simulation-based training (SBT) is gaining popularity as a low-cost and convenient training technique in a vast range of applications. However, for a SBT platform to be fully utilized as an effective training tool, it is essential that feedback on performance is provided automatically in real-time during training. It is the aim of this paper to develop an efficient and effective feedback generation method for the provision of real-time feedback in SBT. Existing methods either have low effectiveness in improving novice skills or suffer from low efficiency, resulting in their inability to be used in real-time. In this paper, we propose a neural network based method to generate feedback using the adversarial technique. The proposed method utilizes a bounded adversarial update to minimize a L1 regularized loss via back-propagation. We empirically show that the proposed method can be used to generate simple, yet effective feedback. Also, it was observed to have high effectiveness and efficiency when compared to existing methods, thus making it a promising option for real-time feedback generation in SBT.","doi":"10.24963/ijcai.2017/526","url":"https://ijcai.org/proceedings/2017/526","pdf":"https://ijcai.org/proceedings/2017/0526.pdf","markdown":"https://ijcai.org/proceedings/2017/526.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/526","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":"Multidisciplinary Topics and Applications","track":"Main track","number":526,"pages":{"first":3763,"last":3769},"keywords":["Machine Learning: Neural Networks","Machine Learning: Data Mining","Multidisciplinary Topics and Applications: Computer-Aided Education"],"videos":[],"references":[{"position":1,"raw":"[Billings, 2012] DR Billings. Efficacy of adaptive feedback strategies in simulation-based training. Military Psychol- ogy, 24(2):114, 2012.","title":"Efficacy of adaptive feedback strategies in simulation-based training","authors":"; Dr Billings, Billings","year":"2012","venue":"Military Psychology","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Chen et al., 2015] Chenyi Chen, Ari Seff, Alain Korn- hauser, and Jianxiong Xiao. Deepdriving: Learning af- fordance for direct perception in autonomous driving. In ICCV, pages 2722-2730, 2015.","title":"Deepdriving: Learning affordance for direct perception in autonomous driving","authors":"Chen","year":"2015","venue":"ICCV","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Cosma and Stanic, 2011] Daniela Cosma and Mircea-Petru Stanic. Implementing a software modeling-simulation in military training. Land Forces Academy Review, 16(2):204, 2011.","title":"Implementing a software modeling-simulation in military training","authors":"Daniela Cosma, Mircea-Petru Stanic","year":"2011","venue":"Land Forces Academy Review","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Cui et al., 2015] Zhicheng Cui, Wenlin Chen, Yujie He, and Yixin Chen. Optimal action extraction for random forests and boosted trees. In KDD, pages 179-188, 2015.","title":"Optimal action extraction for random forests and boosted trees","authors":"Zhicheng Cui, Wenlin Chen, Yujie He, Yixin Chen","year":"2015","venue":"KDD","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Davis, 2005] Walter D Davis. The interactive effects of goal orientation and feedback specificity on task performance. Human Performance, 18(4):409-426, 2005.","title":"The interactive effects of goal orientation and feedback specificity on task performance","authors":"Davis Walter","year":"2005","venue":"Human Performance","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[de Groot et al., 2011] Stefan de Groot, Joost CF de Win- ter, José Manuel López García, Max Mulder, and Peter A Wieringa. The effect of concurrent bandwidth feedback on learning the lane-keeping task in a driving simulator. Human Factors, 53(1):50-62, 2011.","title":"The effect of concurrent bandwidth feedback on learning the lane-keeping task in a driving simulator","authors":"De Groot","year":"2011","venue":"Human Factors","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Forestier et al., 2012] Germain Forestier, Florent Lalys, Laurent Riffaud, Brivael Trelhu, and Pierre Jannin. Classi- fication of surgical processes using dynamic time warping. Journal of Biomedical Informatics, 45(2):255-264, 2012.","title":"Classification of surgical processes using dynamic time warping","authors":"Germain Forestier, Florent Lalys, Laurent Riffaud, Brivael Trelhu, Pierre Jannin","year":"2012","venue":"Journal of Biomedical Informatics","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Glorot et al., 2011] Xavier Glorot, Antoine Bordes, and Yoshua Bengio. Deep sparse rectifier neural networks. In Aistats, volume 15, page 275, 2011.","title":"Deep sparse rectifier neural networks","authors":"Xavier Glorot, Antoine Bordes, Yoshua Bengio","year":"2011","venue":"Aistats","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Goodfellow et al., 2014] Ian J Goodfellow, Jonathon Shlens, and Christian Szegedy. Explaining and harnessing adversarial examples. arXiv:1412.6572, 2014.","title":"Explaining and harnessing adversarial examples","authors":"Ian J Goodfellow, Jonathon Shlens, Christian Szegedy","year":"2014","venue":null,"doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Lateef and others, 2010] Fatimah Lateef et al. Simulation- based learning: Just like the real thing. Journal of emer- gencies, trauma, and shock, 3(4):348, 2010.","title":"Simulationbased learning: Just like the real thing","authors":"Fatimah Lateef, Lateef","year":"2010","venue":"Journal of emergencies, trauma, and shock","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Ma et al., 2017] Xingjun Ma, Sudanthi Wijewickrema, Yun Zhou, Bridget Copson, James Bailey, Gregor Kennedy, and Stephen OLeary. Simulation for training cochlear im- plant electrode insertion. 2017.","title":"Simulation for training cochlear implant electrode insertion","authors":"Ma","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Pascanu et al., 2012] Razvan Pascanu, Tomas Mikolov, and Yoshua Bengio. Understanding the exploding gradi- ent problem. Computing Research Repository (CoRR) abs/1211.5063, 2012.","title":"Understanding the exploding gradient problem","authors":"Razvan Pascanu, Tomas Mikolov, Yoshua Bengio","year":"2012","venue":null,"doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Rhienmora et al., 2011] Phattanapon Rhienmora, Peter Haddawy, Siriwan Suebnukarn, and Matthew N Dailey. Intelligent dental training simulator with objective skill assessment and feedback. Artificial intelligence in medicine, 52(2):115-121, 2011.","title":"Intelligent dental training simulator with objective skill assessment and feedback","authors":"Rhienmora","year":"2011","venue":"Artificial intelligence in medicine","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Rojas et al., 2014] David Rojas, Faizal Haji, Rob Shewaga, Bill Kapralos, and Adam Dubrowski. The impact of secondary-task type on the sensitivity of reaction-time based measurement of cognitive load for novices learning surgical skills using simulation. Stud Health Technol In- form, 196:353-359, 2014.","title":"The impact of secondary-task type on the sensitivity of reaction-time based measurement of cognitive load for novices learning surgical skills using simulation","authors":"David Rojas, Faizal Haji, Rob Shewaga, Bill Kapralos, Adam Dubrowski","year":"2014","venue":"Stud Health Technol Inform","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Steadman et al., 2006] Randolph H Steadman, Wendy C Coates, Yue Ming Huang, Rima Matevosian, Baxter R Larmon, Lynne McCullough, and Danit Ariel. Simulation- based training is superior to problem-based learning for the acquisition of critical assessment and management skills. Critical care medicine, 34(1):151-157, 2006.","title":"Simulationbased training is superior to problem-based learning for the acquisition of critical assessment and management skills","authors":"Wendy C Randolph H Steadman, Coates, Ming Yue, Rima Huang, Matevosian, Lynne Baxter R Larmon, Danit Mccullough, Ariel","year":"2006","venue":"Critical care medicine","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Sutton and Barto, 1998] Richard S Sutton and Andrew G Barto. Reinforcement learning: An introduction, vol- ume 1. MIT press Cambridge, 1998.","title":"Reinforcement learning: An introduction","authors":"S Richard, Andrew G Sutton, Barto","year":"1998","venue":null,"doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Sweller, 1988] John Sweller. Cognitive load during prob- lem solving: Effects on learning. Cognitive science, 12(2):257-285, 1988.","title":"Cognitive load during problem solving: Effects on learning","authors":"John Sweller, Sweller","year":"1988","venue":"Cognitive science","doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Szegedy et al., 2013] Christian Szegedy, Wojciech Zaremba, Ilya Sutskever, Joan Bruna, Dumitru Erhan, Ian Goodfellow, and Rob Fergus. Intriguing properties of neural networks. arXiv:1312.6199, 2013.","title":"Intriguing properties of neural networks","authors":"Szegedy","year":"2013","venue":null,"doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Wijewickrema et al., 2016] Sudanthi Wijewickrema, Yun Zhou, James Bailey, Gregor Kennedy, and Stephen O'Leary. Provision of automated step-by-step procedural guidance in virtual reality surgery simulation. In VRST, pages 69-72. ACM, 2016.","title":"Provision of automated step-by-step procedural guidance in virtual reality surgery simulation","authors":"Wijewickrema","year":"2016","venue":"VRST","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Wijewickrema et al., 2017] Sudanthi Wijewickrema, Brid- get Copson, Yun Zhou, Xingjun Ma, Robert Briggs, James Bailey, Gregor Kennedy, and Stephen OLeary. Design and evaluation of a virtual reality simulation module for train- ing advanced temporal bone surgery. 2017.","title":"Design and evaluation of a virtual reality simulation module for training advanced temporal bone surgery","authors":"Wijewickrema","year":"2017","venue":null,"doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Yang et al., 2003] Qiang Yang, Jie Yin, Charles X Ling, and Tielin Chen. Postprocessing decision trees to extract ac- tionable knowledge. In ICDM, pages 685-688, 2003.","title":"Postprocessing decision trees to extract actionable knowledge","authors":"Qiang Yang, Jie Yin, Charles X Ling, Tielin Chen","year":"2003","venue":"ICDM","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Yang et al., 2007] Qiang Yang, Jie Yin, Charles Ling, and Rong Pan. Extracting actionable knowledge from decision trees. TKDE, 19(1):43-56, 2007.","title":"Extracting actionable knowledge from decision trees","authors":"Qiang Yang, Jie Yin, Charles Ling, Rong Pan","year":"2007","venue":"TKDE","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Zhou et al., 2013a] Yun Zhou, James Bailey, Ioanna Ioan- nou, Sudanthi Wijewickrema, Gregor Kennedy, and Stephen OLeary. Constructive real time feedback for a temporal bone simulator. In MICCAI, pages 315-322. 2013.","title":"Constructive real time feedback for a temporal bone simulator","authors":"Zhou","year":"2013","venue":"MICCAI","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Zhou et al., 2013b] Yun Zhou, James Bailey, Ioanna Ioan- nou, Sudanthi Wijewickrema, Stephen O'Leary, and Gre- gor Kennedy. Pattern-based real-time feedback for a tem- poral bone simulator. In VRST, pages 7-16, 2013.","title":"Pattern-based real-time feedback for a temporal bone simulator","authors":"Zhou","year":"2013","venue":"VRST","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Zhou et al., 2015] Yun Zhou, Ioanna Ioannou, Sudanthi Wi- jewickrema, James Bailey, Gregor Kennedy, and Stephen OLeary. Automated segmentation of surgical motion for performance analysis and feedback. In MICCAI, pages 379-386. 2015.","title":"Automated segmentation of surgical motion for performance analysis and feedback","authors":"Yun Zhou, Ioanna Ioannou, Sudanthi Wijewickrema, James Bailey, Gregor Kennedy, Stephen Oleary","year":"2015","venue":"MICCAI","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}