Modeling the Physical World: Perception, Learning, and Control
NeurIPS Workshop, December 7, 2018, Montreal, Canada
Accepted Papers
Oral Presentations
ChainQueen: A Real-Time Differentiable Physical Simulator for Soft Robotics Yuanming Hu (MIT CSAIL)*; Jiancheng Liu (Tsinghua Univestity); Andrew Spielberg (MIT); Josh Tenenbaum (MIT); Bill Freeman (MIT); Jiajun Wu (MIT); Daniela Rus (MIT CSAIL); Wojciech Matusik (MIT)
To Stir or Not to Stir: Online Estimation of Liquid Properties for Pouring Actions Tatiana Lopez Guevara (University of Edinburgh, Heriot-Watt University)*; Rita Pucci (University of Edinburgh); Nicholas K Taylor (Heriot-Watt University); Michael U. Gutmann (University of Edinburgh); Subramanian Ramamoorthy (University of Edinburgh); Kartic Subr (University of Edinburgh)
Learning Robotic Manipulation through Visual Planning and Acting Angelina Wang (UC Berkeley)*; Thanard Kurutach (UC Berkeley); Aviv Tamar (UC Berkeley); Pieter Abbeel (UC Berkeley)
Posters
Perceiving Physical Equation by Observing Visual Scenarios Siyu Huang (Zhejiang University)*; Zhi-Qi Cheng (Carnegie Mellon University); Xi Li (Zhejiang University); Xiao Wu (Southwest Jiaotong University); Zhongfei Zhang (Zhejiang University); Alexander Hauptmann (Carnegie Mellon University)
Learning Dexterous In-Hand Manipulation Marcin Andrychowicz (OpenAI)*; Bowen Baker (OpenAI); Maciej Chociej (OpenAI); Rafal Jozefowicz (OpenAI); Bob McGrew (OpenAI); Jakub Pachocki (OpenAI); Arthur Petron (OpenAI); Matthias Plappert (OpenAI); Glenn Powell (OpenAI); Alex Ray (OpenAI); Jonas Schneider (OpenAI); Szymon Sidor (OpenAI); Joshua P Tobin (OpenAI); Peter Welinder (OpenAI); Lilian Weng (OpenAI); Wojciech Zaremba (OpenAI)
Transferring Physical Motion Between Domains for Neural Inertial Tracking Changhao Chen (University of Oxford)*; Yishu Miao (University of Oxford); Chris Xiaoxuan Lu (University of Oxford); Phil Blunsom (); Andrew Markham (University of Oxford); Niki Trigoni (University of Oxford)
Learning Tasks in a Complex Circular Maze Environment Jeroen Vanbaar (MERL)*; Devesh K Jha (MERL); Diego Romeres (MERL); Alan Sullivan (MERL); Daniel Nikovski ()
Towards Natural and Accurate Future Pose Prediction for Human and Animals Shuang Wu (Nanyang Technological University)*; Zhenguang Liu (Zhejiang Gongshang University); Shuyuan Jin (NUS); Qi Liu (National University of Singapore); Shijian Lu (Nanyang Technological University); Roger Zimmermann (NUS); Li Cheng ("Bioinformatics Institute, A*STAR, Singapore")
Consistent Jumpy Predictions for Videos and Scenes Ananya Kumar (DeepMind)*; S. M. Ali Eslami (DeepMind); Danilo Jimenez Rezende (Google DeepMind); Marta Garnelo (DeepMind); Fabio Viola (DeepMind); Edward Lockhart (DeepMind); Murray P Shanahan ()
Gray-box probabilistic occupancy mapping Anthony P Tompkins (The University of Sydney)*; Ransalu Senanayake (University of Sydney); Fabio Ramos (U Sydney)
Imagining hidden supporting objects in tabletop scenes Hector Basevi (University of Birmingham)*; Ales Leonardis (University of Birmingham)
Discovering physical concepts with neural networks Raban Iten (ETH Zürich)*; Tony Metger (ETH Zürich); Henrik Wilming (ETH Zürich); Lidia del Rio (ETH Zürich); Renato Renner (ETH Zürich)
Spatially Invariant Attend, Infer, Repeat Eric Crawford (McGill University)*; Joelle Pineau (McGill / Facebook)
Neural Lander: Stable Drone Landing Control using Learned Dynamics Guanya Shi (California Institute of Technology)*; Xichen Shi (Caltech); Michael O'Connell (Caltech); Rose Yu (Northeastern); Kamyar Azizzadenesheli (University of California, Irvine); Animashree Anandkumar (Caltech); Yisong Yue (Caltech); Soon-Jo Chung (Caltech)
Inverse Optimal Power Flow: Assessing theVulnerability of Power Grid Data Priya L Donti (Carnegie Mellon University)*; Zico Kolter (Carnegie Mellon University); Inês Azevedo (Carnegie Mellon University)
Object-Oriented Dynamics Learning through Multi-Level Abstraction Guangxiang Zhu (Tsinghua university)*; Jianhao Wang (Tsinghua University); Zhizhou Ren (Tsinghua University); Chongjie Zhang (Tsinghua University)
Physics-Based Deep Learning for Fluid Flow Nils Thuerey (TUM)*
Combining Analytical and Learned Models for Model Predictive Control Thomas Baumeister (Stanford University)*; Jeannette Bohg (Stanford); Alina Kloss (Max Planck Institute for Intelligent Systems)
Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids Yunzhu Li (MIT)*; Jiajun Wu (MIT); Russ Tedrake (MIT); Joshua Tenenbaum (MIT); Antonio Torralba (MIT)
Learning the Intuitive Physics of Non-Rigid Object Deformations Stefano Rosa (University of Oxford)*; Zhihua Wang (University of Oxford); Andrew Markham (University of Oxford)
Learning Physics with Neural Stethoscopes Fabian Fuchs (Oxford Robotics Insitute)*; Oliver M Groth (Oxford Robotics Insitute); Adam Kosiorek (Oxford); Alex Bewley (University of Oxford); Markus Wulfmeier (Oxford); Andrea Vedaldi (Oxford University); Ingmar Posner (Oxford)
Expert Guided Machine Learning with physical model for adaptive representation of smart grids Antoine Marot (RTE France)*; Sami Tazi (RTE)
Omni-push: accurate, diverse, real-world dataset of pushing dynamics with RGBD images Maria Bauza Villalonga (MIT)*; Ferran Alet (MIT); Tomas Lozano-Perez (MIT); Alberto Rodriguez (MIT); Leslie Kaelbling (MIT)
Incorporating Attention in World Models for Improved Dynamics Modeling Deepika Bablani (IBM Research)*; Parth Chadha (Nvidia)
A Case for Object Compositionality in GANs Sjoerd van Steenkiste (IDSIA)*; Karol Kurach (Google); Sylvain Gelly (Google Brain)
Leveraging Physical Models for Gentle Manipulation Sandy Huang (UC Berkeley)*; Martina Zambelli (Google DeepMind); Yuval Tassa (DeepMind); Jackie Kay (DeepMind); Murilo Martins (DeepMind); Patrick Pilarski (); Raia Hadsell (Google Deepmind)
Towards Bridging Human and Artificial Cognition: Hybrid Variational Predictive Coding of the Physical World, the Body and the Brain André Ofner (University of Potsdam)*
Algorithmic Framework for Model-based Deep Reinforcement Learning with Theoretical Guarantees Yuping Luo (Princeton University)*; Huazhe Xu (UC Berkeley); Yuanzhi Li (Stanford University); Yuandong Tian (Facebook); Trevor Darrell (UC Berkeley); Tengyu Ma (Facebook/Stanford)
Approximating the solution to wave propagation using deep neural networks Wilhelm Sorteberg (Imperial College London); Stef Garasto (Imperial College London)*; Alison Pouplin (Imperial College London); Chris Cantwell (Imperial College London); Anil Anthony Bharath ((Imperial College of London,UK))
Value constrained model-free continuous control Steven Bohez (DeepMind)*; Abbas Abdolmaleki (Google DeepMind); Michael Neunert (Google DeepMind); Jonas Buchli (Google); Nicolas Heess (DeepMind); Raia Hadsell (Google Deepmind)
State representation learning with recurrent capsule networks Louis Annabi (Softbank Robotics Europe)*; Michael Garcia Ortiz (SoftBank Robotics Europe)
Extracting Interpretable Physical Parameters from Partial Differential Equations using Unsupervised Learning Peter Lu (MIT)*; Samuel Kim (MIT); Marin Soljacic (MIT)
Observational and self learning of multi-step robotic manipulation with unknown physical properties Claudia Perez D'Arpino (MIT)*; Julie A. Shah (MIT)
Reasoning About Physical Interactions with Object-Centric Models Michael Janner (UC Berkeley)*; Sergey Levine (UC Berkeley); Bill Freeman (MIT); Joshua Tenenbaum (MIT); Chelsea Finn (UC Berkeley); Jiajun Wu (MIT)
Combining Physical Simulators and Object-Based Networks for Prediction and Control Anurag Ajay (MIT)*; Jiajun Wu (MIT); Maria Bauza Villalonga (MIT); Nima Fazeli (MIT); Joshua Tenenbaum (MIT); Alberto Rodriguez (MIT); Leslie Kaelbling (MIT)
Which resampling methods can tame ill-behaved gradients in chaotic systems? Paavo Parmas (Okinawa Inst. of Sci. and Tech)*; Jan Peters (TU Darmstadt + Max Planck Institute for Intelligent Systems); Kenji Doya (Okinawa Institute of Science and Technology)