Full program
Monday, November 13th
0800 - 0845 - Registration
0845 - 0900 - Welcome & Logistics
0900 - 1000 - Keynote 
                        Rodney Brooks, MIT
1000 - 1020 - Learning Human Utility from Video Demonstrations for Deductive Planning in Robotics
                        Nishant Shukla, UCLA; Song-Chun Zhu, UCLA; Frank Chen, UCLA; Yunzhong He, UCLA
1020 - 1040 - Learning a Visuomotor Controller for Real World Robotic Grasping Using Simulated Depth                            Images
                        Ulrich Viereck, Northeastern University; Andreas ten Pas, Northeastern University; Kate Saenko,                                    Boston University; Robert Platt, Northeastern University
1040 - 1100 - One-Shot Visual Imitation Learning via Meta-Learning 
                        Chelsea Finn, UC Berkeley; Tianhe Yu, UC Berkeley; Tianhao Zhang, UC Berkeley; Pieter Abbeel, UC                                Berkeley; Sergey Levine, UC Berkeley
1100 - 1130 - Coffee Break
1130 - 1136 - Learning Partially Contracting Dynamical Systems from Demonstrations
                        Harish Chaandar Ravichandar, University of Connecticut; Iman Salehi, University of Connecticut;                                    Ashwin Dani, University of Connecticut
1136 - 1142 - Learning Data-Efficient Rigid-Body Contact Models: Case Study of Planar                                                           Impact 
                        Nima Fazeli, MIT; Samuel Zapolsky; Evan Drumwright; Alberto Rodriguez, MIT
1142 - 1148 - Transferring End-to-End Visiomotor Control from Simulation to Real World for a Multi-                                   Stage Task
                        Stephen James, Imperial College London; Andrew Davison, Imperial College London; Edward Johns,                                Imperial College London
1148 - 1154 - Learning Deep Policies for Robot Bin Picking using Discrete-Event Simulation of Robust                                Grasping Sequences
                         Jeffrey Mahler, UC Berkeley; Ken Goldberg, UC Berkeley
1154 - 1200 - Fast Residual Forests: Rapid Ensemble Learning for Semantic Segmentation
                         Yan Zuo, Monash University; Tom Drummond, Monash University
1200 - 1206 - Adaptable Pouring: Teaching Robots Not to Spill using Fast but Approximate Fluid                                         Simulation
                         Tatiana López Guevara, University of Edinburgh & Heriot-Watt University; Nicholas K. Taylor, Heriot-                            Watt University; Michael U. Gutmann, University of Edinburgh; Subramanian Ramamoorthy, University                          of Edinburgh; Kartic Subr, University of Edinburgh
1206 - 1212 - Improved Adversarial Systems for 3D Object Generation and Reconstruction
                         Edward Smith, McGill; David Meger, University of British Columbia
1212 - 1218 - Optimizing Long-term Predictions for Model-based Policy Search
                         Andreas Doerr, MPI-IS, BCAI; Christian Daniel, ; Duy Nguyen-Tuong, ; Alonso Marco, ; Stefan Schaal, ;                             Marc Toussaint, ; Sebastian Trimpe, MPI for Intelligent Systems
1218 - 1224 - Learning Robotic Manipulation of Granular Media
                         Connor Schenck, University of Washington; Sergey Levine, UC Berkeley; Jonathan Tompson, Google;                                 Dieter Fox, University of Washington
1224 - 1230 - Predictive-State Decoders: Augmenting Recurrent Networks for Better Filtering, Imitation,                         and Reinforcement Learning
                         Arun Venkatraman, Carnegie Mellon University; Nick Rhinehart, Carnegie Mellon University; Wen Sun,                             Carnegie Mellon University; Byron Boots, Georgia Institute of Technology; Kris Kitani, Carnegie Mellon                             University; Drew Bagnell, Carnegie Mellon University
1230 - 1330 - Lunch
1330 - 1430 - Keynote
                         Stefanie Tellex, Brown University
1430 - 1450 - Opportunistic Active Learning for Grounding Natural Language Descriptions
                          Jesse Thomason, University of Texas at Austin; Aishwarya Padmakumar, University of Texas at Austin;                           Jivko Sinapov, University of Texas at Austin; Justin Hart, University of Texas at Austin; Peter Stone,                                 University of Texas at Austin; Raymond Mooney, University of Texas at Austin
1450 - 1510 - Towards Robust Skill Generalization: Unifying Learning from Demonstration and Motion                             Planning
                          Muhammad Asif Rana, Georgia Institute of Technology; Mustafa Mukadam, Georgia Tech; Seyed Reza                           Ahmadzadeh, Georgia Institute of Technology; Sonia Chernova, Georgia Institute of Technology; Byron                           Boots, Georgia Institute of Technology
1510 - 1530 - Learning Robot Objectives from Physical Human Interaction
                         Andrea Bajcsy, UC Berkeley; Dylan Losey, Rice University; Marcia O'Malley, Rice University; Anca                                      Dragan, UC Berkeley
1530 - 1600 - Coffee Break
1600 - 1606 - Active Incremental Learning of Robot Movement Primitives
                         Guilherme Maeda, TU Darmstadt; Marco Ewerton, TU Darmstadt; Takayuki Osa, University of Tokyo;                             Baptiste Busch, Inria-Bordeaux; Jan Peters, TU Darmstadt
1606 - 1612 - Dart: Optimizing Noise Injection in Imitation Learning
                         Michael Laskey, UC Berkeley; Anca Dragan, UC Berkeley; Jonathan Lee, UC Berkeley; Ken Goldberg, UC                             Berkeley; Roy Fox, UC Berkeley
1612 - 1618 - Bayesian Interaction Primitives: A SLAM Approach to Human-Robot Interaction
                         Joseph Campbell, Arizona State University; Heni Ben Amor, Arizona State University
1618 - 1624 - Hierarchical Reinforcement Learning with Parameters
                         Piotr Milos, University of Warsaw; Henryk Michalewski, University of Warsaw; Maciej Klimek,                                   
1624 - 1630 - DDCO: Discovery of Deep Continuous Options for Robot Learning from Demonstrations
                         Sanjay Krishnan, UC Berkeley; Roy Fox, UC Berkeley; Ion Stoica, UC Berkeley; Ken Goldberg, UC Berkeley
1630 - 1636 - Extending Model-based Policy Gradients for Robots in Heteroscedastic Environments
                         John Martin, Stevens Institute of Technolog; Brendan Englot, Stevens Institute of Technology
1636 - 1642 - Neural Task Programming: Learning to Generalize Across Hierarchical Tasks
                         Danfei Xu, Stanford University; Yuke Zhu,Stanford University; Yuan Gao, Stanford University; Animesh                             Garg, Stanford University; Li Fei-Fei, Stanford University; Silvio Savarese, Stanford University
1642 - 1648 - Most Likely Expected Improvement for Automatic Prior Selection in Data-Efficient Direct                             Policy Search
                         Rémi Pautrat, Inria Nancy - Grand-Est; Konstantinos Chatzilygeroudis, Inria Nancy Grand-Est; Jean-                                Baptiste Mouret, INRIA
1648 - 1654 - Learning Time Invariant Driver Styles with Burn-InfoGAIL
                         Alex Kuefler, Stanford University; Mykel Kochenderfer, Stanford University
1654 - 1700 - Action Learning for Multi-Agent Navigation
                          Julio Godoy, Universidad de Concepción; Tiannan Chen, University of Minnesota; Stephen Guy,                                     University of Minnesota; Ioannis Karamouzas, Clemson University; Maria Gini, University of Minnesota
1700 - 1800 - Poster session
1800 - 2000 - Dinner at the Google Campus

Tuesday, November 14th
0900 - 1000 - Keynote 
                        Yann LeCun, Facebook & NYU
1000 - 1020 - image2mass: Estimating the Mass of an Object from Its Image 
                        Trevor Standley, Stanford University; Ozan Sener, Stanford University; Silvio Savarese, Stanford                                        University
1020 - 1040 - Learning Stable Task Sequences from Demonstration with Linear Parameter Varying                                    Systems and Hidden Markov Models
                         Jose Medina, EPFL; Aude Billard, EPFL
1040 - 1100 - Principal Variety Analysis
                         Reza Iraji, Colorado State University; Hamidreza Chitsaz, Colorado State University
1100 - 1130 - Coffee Break
1130 - 1136 - CARLA: An Open Urban Driving Simulator
                         Alexey Dosovitskiy, Intel Labs; German Ros, Computer Vision Center; Felipe Codevilla, Computer Vision                          Center; Antonio Lopez, Computer Vision Center (CVC); Vladlen Koltun, Intel Labs
1136 - 1142 - Aggressive Deep Driving: Combining Convolutional Neural Networks and Model Predictive                         Control
                         Paul Drews, Georgia Institute of Technology; Grady Williams, Georgia Institute of Technology; Brian                                 Goldfain, Georgia Institute of Technology; Evangelos Theodorou, Georgia Institute of Technology; James                          Rehg, Georgia Institute of Technology
1142 - 1148 - Intention-Net: Integrated Planning and Deep Learning for Autonomous Navigation
                         Wei Gao, NUS; Karthikk Subramanian, Panasonic R&D
1148 - 1154 - Gradient-free Policy Architecture Search and Adaptation
                         Sayna Ebrahimi, UC Berkeley; Anna Rohrbach, Max Plank Institute for Informatics; Trevor Darrell, UC                             Berkeley
1154 - 1200 - Long-Term On-Board Prediction of Pedestrians in Traffic Scenes
                         Apratim Bhattacharyya, MPI Informatics; Mario Frtiz, MPI Informatics; Bernt Schiele
1200 - 1206 - Semi-Supervised Haptic Material Recognition for Robots using Generative Adversarial                                 Networks
                         Zackory Erickson, Georgia Institute of Technology; Sonia Chernova, Georgia Institute of Technology;                                 Charles Kemp, Georgia Institute of Technology
1206 - 1212 - ICORe50: a New Dataset and Benchmark for Continuous Object Recognition
                         Vincenzo Lomonaco, University of Bologna; davide Maltoni, University of Bologna
1212 - 1218 - How Robots Learn to Classify New Objects Trained from Small Data Sets
                         Tick Son Wang, DLR; Zoltan-Csaba Marton, German Aerospace Center (DLR); Manuel Brucker, DLR;                                 Rudolph Triebel
1218 - 1224 - Learning to Fly by Crashing
                         Dhiraj Gandhi, Carngie Mellon University; Lerrel Pinto, Carngie Mellon University; Abhinav Gupta,                                 Carngie Mellon University
1224 - 1230 - Towards Grasp Transfer using Shape Deformation
                         Andrey Kurenkov, Stanford University; Viraj Mehta, Stanford University; Jingwei Ji, Stanford University;                             Animesh Garg, Stanford University; Silvio Savarese, Stanford University
1230 - 1330 - Lunch
1330 - 1430 - Keynote
                         Drew Bagnell, Carnegie Mellon University
1430 - 1450 - Sim-to-Real Robot Learning from Pixels with Progressive Nets
                         Andrei Rusu, DeepMind; Matej Vecerik, DeepMind; Thomas Rothorl, DeepMind; Nicolas Heess,                                         DeepMind; Razvan Pascanu, DeepMind; Raia Hadsell, Google DeepMind
1450 - 1510 - Learning Heuristic Search via Imitation
                         Mohak Bhardwaj, Carnegie Mellon University; Sanjiban Choudhury, Carnegie Mellon University;                                     Sebastian Scherer, Carnegie Mellon University
1510 - 1530 - Mutual Alignment Transfer Learning
                         Markus Wulfmeier, Oxford; Ingmar Posner, Oxford; Pieter Abbeel, UC Berkeley
1530 - 1600 - Coffee Break
1600 - 1606 - Uncertainty-driven Imagination for Continuous Deep Reinforcement Learning
                        Gabriel Kalweit, University of Freiburg; Joschka Boedecker, University of Freiburg
1606 - 1612 - The Intentional Unintentional Agent:Learning to Solve Many Continuous Control Tasks                                Simultaneously
                        Serkan Cabi, DeepMind; Sergio Gomez Colmenarejo, DeepMind; Matt Hoffman, DeepMind; Misha Denil,                         DeepMind; Ziyu Wang, DeepMind; Nando de Freitas, DeepMind
1612 - 1618 - Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation
                        Guan-Horng Liu, Carnegie Mellon University; Avinash Siravuru, Carnegie Mellon University; Sai                                        Prabhakar, Carnegie Mellon University; Manuela Veloso, Carnegie Mellon University; George Kantor,                            Carnegie Mellon University
1618 - 1624 - Reverse Curriculum Generation for Robotic Manipulation with Reinforcement Learning
                        Carlos Florensa, UC Berkeley; David Held, UC Berkeley; Pieter Abbeel, UC Berkeley
1624 - 1630 - Trust-PCL: An Off-Policy Trust Region Method for Continuous Control
                        Ofir Nachum, Google; Mohammad Norouzi, Google; Kelvin Xu, Google; Dale Schuurmans, Google
1630 - 1636 - IntentionGAN: Multi-Task Imitation Learning from Unstructured Demonstrations
                         Karol Hausman, USC; Yevgen Chebotar, USC; Stefan Schaal, USC; Gaurav Sukhatme, USC; Joseph Lim,                             USC
1636 - 1642 - Manifold Regularization for Kernelized LSTD
                         Xinyan Yan, Google; Krzysztof Choromanski, ; Byron Boots, Georgia Institute of Technology; Vikas                                     Sindhwani, Google
1642 - 1648 - Efficient Automatic Perception System Parameter Tuning On Site without Expert                                            Supervision
                         Humphrey Hu, Carnegie Mellon University; George Kantor, Carnegie Mellon University
1648 - 1654 - Predictive State Models for Prediction and Control in Partially Observable Environments
                         Ahmed Hefny, Carnegie Mellon University; Zita Marinho, Carnegie Mellon University; Wen Sun,                                         Carnegie Mellon University; Carlton Downey, Carnegie Mellon University; Goeffrey Gordon, Carnegie                              Mellon University
1654 - 1700 - Predictive State Recurrent Neural Networks
                         Ahmed Hefny, Carnegie Mellon University; Carlton Downey, Carnegie Mellon University; Byron Boots,                             Georgia Institute of Technology; Goeffrey Gordon, Carnegie Mellon University
1700 - 1800 - Poster session
1800 - 2100 - Reception at the Computer History Museum

Wednesday, November 15th
0900 - 1000 - Keynote 
                        Anca Dragan, UC Berkeley 
1000 - 1020 - Occlusion-Aware Visual Foresight for Self-Supervised Robot Learning
                        Frederik Ebert, UC Berkeley; Chelsea Finn, UC Berkeley; Alex Lee, UC Berkeley; Sergey Levine, UC Berkeley
1020 - 1040 - Learning Dynamics Across Similar Spatiotemporally Evolving Systems
                        Joshua Whitman, UIUC; Girish Chowdhary, UIUC
1040 - 1100 - Fastron: An Online Learning-Based Model and Active Learning Strategy for Proxy Collision                            Detection
                        Nikhil Das, UC San Diego; Naman Gupta, UC San Diego; Michael Yip, UC San Diego
1100 - 1130 - Coffee Break
1130 - 1136 - End-to-End Learning of Semantic Grasping
                        Eric Jang, Google; Sudheendra Vijayanarasimhan,; Peter Pastor, [X]; Julian Ibarz, Google;                            Sergey Levine, UC Berkeley
1136 - 1142 - Deep Neural Networks as Add-on Modules for High-Accuracy Impromptu Trajectory                                    Tracking
                         SiQi Zhou, University of Toronto; Mohamed Helwa, University of Toronto; Angela Schoellig, University of                          Toronto
1142 - 1148 - Bayesian Hilbert Maps for Dynamic Continuous Occupancy Mapping
                        Ransalu Senanayake, The University of Sydney; Fabio Ramos, The University of Sydney
1148 - 1154 - Harvesting Common-Sense Navigational Knowledge for Robotics from Uncurated Text                                Corpora

                        Nancy Fulda, BYU PCCL; Zachary Brown, BYU PCCL; Nathan Tibbetts, BYU PCCL
1154 - 1200 - QMDP-Net: Deep Learning for Planning under Partial Observability
                        Peter Karkus, National University of Singapo; David Hsu, NUS; Wee Sun Lee, NUS
1200 - 1206 - Learning Deep Grasping Models From Vision and Touch
                        Roberto Calandra, UC Berkeley; Andrew Owens, UC Berkeley; Manu Upadhyaya, UC Berkeley; Wenzhen                         Yuan, MIT; Justin Lin, UC Berkeley; Edward Adelson, MIT; Sergey Levine, UC Berkeley
1206 - 1212 - Bodily Aware Soft Robots: Integration of Proprioceptive and Exteroceptive Sensors
                        Gabor Soter, University of Bristol; Jonathan Rossiter, University of Bristol; Helmut Hauser, University of                         Bristol; Andrew Conn, University of Bristol
1212 - 1218 - Deep Kernels for Optimizing Locomotion Controllers
                        Rika Antonova, KTH; Akshara Rai, Carnegie Mellon University; Christopher Atkeson, Carnegie Mellon                                University
1218 - 1224 - Emergent Behaviors in Mixed-Autonomy Traffic
                        Cathy Wu, UC Berkeley; Abdul Kreidieh, UC Berkeley; Eugene Vinitsky, UC Berkeley; Alexandre Bayen, UC                         Berkeley
1224 - 1230 - Seeing the Force: Integrating Poses and Visually Latent Forces forManipulations through                             Fluent Discovery and Imitation Learning
                        Feng Gao; Mark Edmonds, ; Xu Xie, UCLA; Hangxin Liu, University of California, Los ; Siyuan Qi, UCLA;                            Yixin Zhu, UCLA; Brandon Rothrock; Song-Chun Zhu, UCLA
1230 - 1330 - Lunch
1330 - 1350 - Online Learning with Stochastic Recurrent Neural Networks using Intrinsic Motivation                                 Signals
                         Daniel Tanneberg, TU Darmstadt; Jan Peters, TU Darmstadt; Elmar Rueckert, TU Darmstadt
1350 - 1410 - MBMF: Model-Based Priors for Model-Free Reinforcement Learning
                         Somil Bansal, UC Berkeley; Roberto Calandra, UC Berkeley; Sergey Levine, UC Berkeley; Claire Tomlin,                             UC Berkeley
1410 - 1430 - Safe Model-based Reinforcement Learning with Stability Guarantees
                         Felix Berkenkamp, ETH Zurich; Matteo Turchetta, ETH Zurich; Angela Schoellig, University of Toronto;                             Andreas Krause, ETH Zurich
1430 - 1530 - Open Questions
                         Ken Goldberg, Pieter Abbeel, Raia Hadsell, Leslie Kaebling
1530 - 1630 - Poster session

1530 - 1600 - Envoi
                         Vincent Vanhoucke, Sergey Levine

Program at a glance

Keynote speakers

Drew Bagnell | Associate Professor, Carnegie Mellon University
J. A. (Drew) Bagnell is an associate professor at Carnegie Mellon University’s Robotics Institute and Machine Learning Department. His interests in artificial intelligence range from algorithmic and basic theoretical development to delivering fielded learning-based systems. Bagnell directs the Learning, AI, and Robotics Laboratory (LAIRLab) within the Robotics Institute. In the past two years, Dr. Bagnell has been on sabbatical to lead efforts on perception and autonomy software efforts for self-driving vehicles.

Bagnell and his group have received over a dozen "best paper" research awards in both the robotics and machine learning communities including at the International Conference on Machine Learning, Robotics Science and Systems, and International Conference on Robotics and Automation. Bagnell received the 2016 Ryan Award, Carnegie Mellon’s award for meritorious teaching, and has served as the founding director of the Robotics Institute Summer Scholars program, a summer research experience that has enabled hundreds of undergraduates throughout the world to leap into robotics research.

His robotics focus includes machine learning for dexterous manipulation, decision making under uncertainty, agile ground and aerial vehicle control, robot perception and computer vision.

Rodney Brooks | Professor Emeritus, MIT
A mathematics undergraduate in his native Australia, Rodney received a Ph.D. in Computer Science from Stanford in 1981. From 1984 to 2010, he was on the MIT faculty, and completed his service as a Professor of Robotics. He was also the founding Director of the Institute’s Computer Science and Artificial Intelligence Laboratory, and served in that role until 2007. In 1990, he co-founded iRobot (NASDAQ: IRBT), where he served variously as CTO, Chairman and board member until 2011. Rodney has been honored by election to the National Academy of Engineering, and has been elected as a Fellow of the American Academy of Arts and Sciences, the Association of Computing Machinery, the Association for the Advancement of Artificial Intelligence, the Institute of Electrical and Electronics Engineers and the American Association for the Advancement of Science. Rodney is also an accomplished presenter, and speaks regularly to promote the value of robotics and artificial intelligence in venues throughout the world.

Anca Dragan | Assistant Professor, UC Berkeley
Autonomy that interacts - that collaborates and coexists with humans - is becoming more and more functional, making its way out of research labs and into industry. Anca's goal is to weave interaction into the very fabric of this autonomy. She envisions riding in a self-driving car as it is effectively coordinating with other drivers on the road and with pedestrians. She envisions people with disabilities seamlessly operating assistive devices to thrive independently. And she envisions collaborative robots in the home or in the factory helping us with our tasks and even gently guiding us to better ways of achieving them.

Anca completed her Ph.D. at Carnegie Mellon in the Robotics Institute. She runs the InterACT Lab, serves on the steering committee for the Berkeley AI Research (BAIR) Lab, and is a co-PI of the Center for Human-Compatible AI.

Yann LeCun | Director of AI Research, Facebook and Silver Professor, NYU
Yann LeCun is Director of AI Research at Facebook and Silver Professor at New York University, affiliated with the Courant Institute, the Center for Neural Science and the Center for Data Science, for which he served as founding director until 2014. He received an EE Diploma from ESIEE (Paris) in 1983, a PhD in Computer Science from Université Pierre et Marie Curie (Paris) in 1987. After a postdoc at the University of Toronto, he joined AT&T Bell Laboratories. He became head of the Image Processing Research Department at AT&T Labs-Research in 1996, and joined NYU in 2003 after a short tenure at the NEC Research Institute. In late 2013, LeCun became Director of AI Research at Facebook, while remaining on the NYU Faculty part-time. He was visiting professor at Collège de France in 2016. His research interests include machine learning and artificial intelligence, with applications to computer vision, natural language understanding, robotics, and computational neuroscience. He is best known for his work in deep learning and the invention of the convolutional network method which is widely used for image, video and speech recognition. He is a member of the US National Academy of Engineering, the recipient of the 2014 IEEE Neural Network Pioneer Award, the 2015 IEEE Pattern Analysis and Machine Intelligence Distinguished Researcher Award, the 2016 Lovie Award for Lifetime Achievement, and a honorary doctorate from IPN, Mexico.

Stefanie Tellex | Assistant Professor, Brown University
Stefanie Tellex is an Assistant Professor of Computer Science and Assistant Professor of Engineering at Brown University. Her group, the Humans To Robots Lab, creates robots that seamlessly collaborate with people to meet their needs using language, gesture, and probabilistic inference, aiming to empower every person with a collaborative robot. She completed her Ph.D. at the MIT Media Lab in 2010, where she developed models for the meanings of spatial prepositions and motion verbs. Her postdoctoral work at MIT CSAIL focused on creating robots that understand natural language. She has published at SIGIR, HRI, RSS, AAAI, IROS, ICAPs and ICMI, winning Best Student Paper at SIGIR and ICMI, Best Paper at RSS, and an award from the CCC Blue Sky Ideas Initiative. Her awards include being named one of IEEE Spectrum's AI's 10 to Watch in 2013, the Richard B. Salomon Faculty Research Award at Brown University, a DARPA Young Faculty Award in 2015, a NASA Early Career Award in 2016, a 2016 Sloan Research Fellowship, and an NSF Career Award in 2017. Her work has been featured in the press on National Public Radio, BBC, MIT Technology Review, Wired and Wired UK, as well as the Smithsonian. She was named one of Wired UK's Women Who Changed Science In 2015 and listed as one of MIT Technology Review's Ten Breakthrough Technologies in 2016.