Timezone: »
Friday, December 07, 2018 at Room 513ABC
Abstract: Understanding the evolution of a process over space and time is fundamental to a variety of disciplines. To name a few, such phenomena that exhibit dynamics in both space and time include propagation of diseases, variations in air pollution, dynamics in fluid flows, and patterns in neural activity. In addition to these fields in which modeling the nonlinear evolution of a process is the focus, there is also an emerging interest in decision-making and controlling of autonomous agents in the spatiotemporal domain. That is, in addition to learning what actions to take, when and where to take actions is crucial for an agent to efficiently and safely operate in dynamic environments. Although various modeling techniques and conventions are used in different application domains, the fundamental principles remain unchanged. Automatically capturing the dependencies between spatial and temporal components, making accurate predictions into the future, quantifying the uncertainty associated with predictions, real-time performance, and working in both big data and data scarce regimes are some of the key aspects that deserve our attention. Establishing connections between Machine Learning and Statistics, this workshop aims at;
(1) raising open questions on challenges of spatiotemporal modeling and decision-making,
(2) establishing connections among diverse application domains of spatiotemporal modeling, and
(3) encouraging conversation between theoreticians and practitioners to develop robust predictive models.
Keywords
Theory: deep learning/convolutional LSTM, kernel methods, chaos theory, reinforcement learning for dynamic environments, dynamic policy learning, biostatistics,
epidemiology, geostatistcs, climatology, neuroscience, etc.
Applications:
Natural phenomena: disease propagation and outbreaks, environmental monitoring, climate modeling, etc.
Social and economics: predictive policing, population mapping, poverty mapping, food resources, agriculture, etc.
Engineering/robotics: active data collection, traffic modeling, motion prediction, fluid dynamics, spatiotemporal prediction for safe autonomous driving, etc.
Web: https://sites.google.com/site/nips18spatiotemporal/
Fri 5:30 a.m. - 6:45 a.m.
|
Christopher Wikle (Uni. of Missouri): Introduction to spatiotemporal modeling
(
Invited talk
)
link »
Christopher K. Wikle is Curators’ Distinguished Professor of Statistics at the University of Missouri (MU), with additional appointments in Soil, Environmental and Atmospheric Sciences and the Truman School of Public Affairs. He received a PhD co-major in Statistics and Atmospheric Science in 1996 from Iowa State University. He was research fellow at the National Center for Atmospheric Research from 1996-1998, after which he joined the MU Department of Statistics. His research interests are in spatio-temporal statistics applied to environmental, geophysical, agricultural and federal survey applications, with particular interest in dynamics. Awards include elected Fellow of the American Statistical Association (ASA), Distinguished Alumni Award from the College of Liberal Arts and Sciences at Iowa State University, ASA ENVR Section Distinguished Achievement Award, co-awardee 2017 ASA Statistical Partnership Among Academe, Industry, and Government (SPAIG) Award, the MU Chancellor’s Award for Outstanding Research and Creative Activity in the Physical and Mathematical Sciences, the Outstanding Graduate Faculty Award, and Outstanding Undergraduate Research Mentor Award. His book Statistics for Spatio-Temporal Data (co-authored with Noel Cressie) was the 2011 PROSE Award winner for excellence in the Mathematics Category by the Association of American Publishers and the 2013 DeGroot Prize winner from the International Society for Bayesian Analysis. He is Associate Editor for several journals and is one of six inaugural members of the Statistics Board of Reviewing Editors for Science. |
Christopher Wikle 🔗 |
Fri 6:45 a.m. - 7:00 a.m.
|
Spotlight talks (session 1)
(
Spotlight talks
)
|
Denisa Roberts · David Kozak · Kehinde Owoeye · astrid dahl · Abdi-Hakin Dirie · Wei-Cheng Chang · Vladimir Ivashkin 🔗 |
Fri 7:00 a.m. - 7:15 a.m.
|
Modeling Rape Reporting Delays Using Spatial, Temporal and Social Features
(
Contributed talk
)
|
Konstantin Klemmer 🔗 |
Fri 7:15 a.m. - 7:30 a.m.
|
Spotlight talks (session 2)
(
Spotlight talks
)
|
Sophie Giffard-Roisin · Marc Rußwurm · Esra Suel · Binh Tang · Harshal Maske · Daniel Neill · Doyup Lee 🔗 |
Fri 7:30 a.m. - 8:00 a.m.
|
Coffee break + poster session 1
(
Coffee break + poster session
)
|
🔗 |
Fri 8:00 a.m. - 8:30 a.m.
|
Stefano Ermon (Stanford University): Weakly Supervised Spatio-temporal Regression
(
Invited talk
)
link »
Stefano Ermon is an Assistant Professor in the Department of Computer Science at Stanford University, where he is affiliated with the Artificial Intelligence Laboratory and a fellow of the Woods Institute for the Environment. His research is centered on techniques for scalable and accurate inference in graphical models, statistical modeling of data, large-scale combinatorial optimization, and robust decision making under uncertainty, and is motivated by a range of applications, in particular ones in the emerging field of computational sustainability with applications in poverty mapping and remote sensing. |
Stefano Ermon 🔗 |
Fri 8:30 a.m. - 8:45 a.m.
|
Long Range Sequence Generation via Multiresolution Adversarial Training
(
Contributed talk
)
|
Rose Yu 🔗 |
Fri 8:45 a.m. - 9:00 a.m.
|
Modeling Spatiotemporal Multimodal Language with Recurrent Multistage Fusion
(
Contributed talk
)
|
Paul Pu Liang 🔗 |
Fri 9:00 a.m. - 9:20 a.m.
|
Spotlight talks (session 3)
(
Spotlight talks
)
|
Farzaneh Mahdisoltani · Frederik Kratzert · SUBBAREDDY OOTA · Mehul Motani · Tryambak Gangopadhyay · Sathwik Tejaswi Madhusudhan · Marc Rußwurm · Mahta Mousavi · Mihir Jain 🔗 |
Fri 9:20 a.m. - 10:45 a.m.
|
Lunch break
|
🔗 |
Fri 10:45 a.m. - 11:00 a.m.
|
A Nonparametric Spatio-temporal SDE Model
(
Contributed talk
)
|
Harri Lahdesmaki 🔗 |
Fri 11:00 a.m. - 11:15 a.m.
|
Spotlight talks (session 4)
(
Spotlight talks
)
|
Anwar Walid · Yitong Li · Ehsan Pajouheshgar · Oliver Hennigh · Seongchan Kim · Vaibhav Kulkarni · Koh Takeuchi 🔗 |
Fri 11:15 a.m. - 11:45 a.m.
|
Ani Hsieh (UPenn): Modeling, Tracking, and Learning Coherent Spatiotemporal Features in Geophysical Flows
(
Invited talk
)
link »
M. Ani Hsieh is a Research Associate Professor in the Department of Mechanical Engineering & Applied Mechanics at the University of Pennsylvania. She received a B.S. in Engineering and B.A. in Economics from Swarthmore College in 1999 and a PhD in Mechanical Engineering from the University of Pennsylvania in 2007. Dr. Hsieh has been a Visiting Assistant Professor in the Engineering Department at Swarthmore College (2007-2008) and an Associate Professor in the Mechanical Engineering & Mechanics Department at Drexel University (2008-2017). Her research interests include many robot systems and marine robotics, geophysical fluid dynamics, and dynamical systems. She is a recipient of a 2012 Office of Naval Research (ONR) Young Investigator Award and a 2013 National Science Foundation (NSF) CAREER Award. |
M. Ani Hsieh 🔗 |
Fri 11:45 a.m. - 12:00 p.m.
|
Spotlight talks (session 5)
(
Spotlight talks
)
|
Alexis Asseman · Roman Marchant · Rakshit Trivedi · Balakrishnan Narayanaswamy · Massinissa AMROUCHE · Henry Martin · Nelson FERNANDEZ PINTO 🔗 |
Fri 12:00 p.m. - 12:30 p.m.
|
Coffee break + poster session 2
(
Coffee break + poster session
)
|
🔗 |
Fri 12:30 p.m. - 1:00 p.m.
|
Chelsea Finn (UCBerkeley / Google Brain): Learning Generalizable Behavior through Unsupervised Interaction
(
Invited talk
)
link »
Chelsea Finn is a postdoctoral fellow in Computer Science at UC Berkeley, where she works on machine learning and its intersection with robotic perception and control. She is a part of Berkeley AI Research Lab (BAIR). She recently spent time at Google Brain. Before graduate school, she received a Bachelors in EECS at MIT, where she worked on several research projects, including an assistive technology project in CSAIL and an animal biometrics project. She has also spent time at Counsyl, Google, and Sandia National Labs. She will discuss on spatiotemporal aspects of video prediction and deep spatial autoencoders for visuomotor learning. |
Chelsea Finn 🔗 |
Fri 1:00 p.m. - 1:30 p.m.
|
Girish Chowdhary (UIUC): Spatiotemporal Learning for Enabling Agricultural Robotics
(
Invited talk
)
link »
Girish Chowdhary is the director of DAS laboratory and Assistant Professor. Girish has a Ph.D. degree from Georgia Institute of Technology. He then spent around two years at Massachusetts Institute of Technology’s Laboratory for Information and Decision Systems and the School of Aeronautics and Astronautics as a postdoctoral associate. Prior to coming to Georgia Tech, he spent three years working as a research engineer with the German Aerospace Center’s (DLR’s) Institute for Flight Systems Technology in Braunschweig, Germany. He holds a BE with honors from RMIT University in Melbourne, Australia. Girish is the author of several peer-reviewed publications spanning the area of adaptive control, spatiotemporal modeling, autonomy and decision making, LIDAR-based perception for Unmanned Aerial Systems (UAS), and GPS denied navigation. |
Girish Chowdhary 🔗 |
Fri 1:30 p.m. - 1:45 p.m.
|
Quantile Regression Reinforcement Learning with State Aligned Vector Rewards
(
Contributed talk
)
|
Oliver Richter 🔗 |
Fri 1:45 p.m. - 2:00 p.m.
|
Path Planning for Mobile Inference of Spatiotemporally Evolving Systems
(
Contributed talk
)
|
Joshua Whitman 🔗 |
Fri 2:00 p.m. - 2:30 p.m.
|
Fabio Ramos (Uni. of Sydney): Learning and Planning in Spatial-Temporal Data
(
Invited talk
)
link »
Abstract: Modern sensors provide immense amounts of information that need to be efficiently integrated into probabilistic models representing the environment autonomous systems operate in. In this talk I will show statistical machine learning methods for spatial and spatial-temporal data that are able to fuse information from heterogeneous sources, scaling gracefully to very large datasets. I will demonstrate how Bayesian reasoning and the principle of modelling uncertainty can be used to mitigate risks in decision making, for motion planning with indoor robots, to continental-scale natural resource exploration. Bio: Fabio Ramos is an Associate Professor in machine learning and robotics at the School of Information Technologies, University of Sydney, and co-Director of the Centre for Translational Data Science. He received the B.Sc. and the M.Sc. degrees in Mechatronics Engineering at University of Sao Paulo, Brazil, in 2001 and 2003 respectively, and the Ph.D. degree at University of Sydney, Australia, in 2008. He has over 130 peer-reviewed publications and received best paper awards at ECML’18, IROS’05, ACRA’07, and Best Paper Finalist at RSS’17. His research focuses on statistical machine learning techniques for data fusion, with applications in robotics, large-scale autonomous systems, environmental monitoring and healthcare. |
Fabio Ramos 🔗 |
Fri 2:30 p.m. - 3:00 p.m.
|
Tomaso Poggio (MIT): Dynamical System Theory for Deep Learning ( Invited talk ) link » | Tomaso Poggio 🔗 |
Fri 3:00 p.m. - 3:30 p.m.
|
Panel Discussion
|
🔗 |
Author Information
Ransalu Senanayake (The University of Sydney)
I am a Computer Science PhD student at the University of Sydney specializing in Machine Learning and Robotics.
Neal Jean (Stanford University)
Fabio Ramos (University of Sydney)
Girish Chowdhary (University of Illinois at Urbana Champaign)
More from the Same Authors
-
2022 : Variance Reduction in Off-Policy Deep Reinforcement Learning using Spectral Normalization »
Payal Bawa · Rafael Oliveira · Fabio Ramos -
2022 : Learning Successor Feature Representations to Train Robust Policies for Multi-task Learning »
Melissa Mozifian · Dieter Fox · David Meger · Fabio Ramos · Animesh Garg -
2022 Workshop: 5th Robot Learning Workshop: Trustworthy Robotics »
Alex Bewley · Roberto Calandra · Anca Dragan · Igor Gilitschenski · Emily Hannigan · Masha Itkina · Hamidreza Kasaei · Jens Kober · Danica Kragic · Nathan Lambert · Julien PEREZ · Fabio Ramos · Ransalu Senanayake · Jonathan Tompson · Vincent Vanhoucke · Markus Wulfmeier -
2022 Spotlight: Batch Bayesian optimisation via density-ratio estimation with guarantees »
Rafael Oliveira · Louis Tiao · Fabio Ramos -
2022 Poster: Batch Bayesian optimisation via density-ratio estimation with guarantees »
Rafael Oliveira · Louis Tiao · Fabio Ramos -
2020 : Invited Talk - "RL with Sim2Real in the Loop / Online Domain Adaptation for Mapping" »
Fabio Ramos · Anthony Tompkins -
2020 : Discussion Panel »
Pete Florence · Dorsa Sadigh · Carolina Parada · Jeannette Bohg · Roberto Calandra · Peter Stone · Fabio Ramos -
2020 : Bayesian optimization by density ratio estimation »
Louis Tiao · Aaron Klein · Cedric Archambeau · Edwin Bonilla · Matthias W Seeger · Fabio Ramos -
2020 Poster: Sparse Spectrum Warped Input Measures for Nonstationary Kernel Learning »
Anthony Tompkins · Rafael Oliveira · Fabio Ramos -
2019 : Poster Session »
Lili Yu · Aleksei Kroshnin · Alex Delalande · Andrew Carr · Anthony Tompkins · Aram-Alexandre Pooladian · Arnaud Robert · Ashok Vardhan Makkuva · Aude Genevay · Bangjie Liu · Bo Zeng · Charlie Frogner · Elsa Cazelles · Esteban G Tabak · Fabio Ramos · François-Pierre PATY · Georgios Balikas · Giulio Trigila · Hao Wang · Hinrich Mahler · Jared Nielsen · Karim Lounici · Kyle Swanson · Mukul Bhutani · Pierre Bréchet · Piotr Indyk · samuel cohen · Stefanie Jegelka · Tao Wu · Thibault Sejourne · Tudor Manole · Wenjun Zhao · Wenlin Wang · Wenqi Wang · Yonatan Dukler · Zihao Wang · Chaosheng Dong -
2018 : Fabio Ramos (Uni. of Sydney): Learning and Planning in Spatial-Temporal Data »
Fabio Ramos -
2018 : Girish Chowdhary (UIUC): Spatiotemporal Learning for Enabling Agricultural Robotics »
Girish Chowdhary -
2018 Poster: Integrated accounts of behavioral and neuroimaging data using flexible recurrent neural network models »
Amir Dezfouli · Richard Morris · Fabio Ramos · Peter Dayan · Bernard Balleine -
2018 Poster: Semi-supervised Deep Kernel Learning: Regression with Unlabeled Data by Minimizing Predictive Variance »
Neal Jean · Sang Michael Xie · Stefano Ermon -
2018 Oral: Integrated accounts of behavioral and neuroimaging data using flexible recurrent neural network models »
Amir Dezfouli · Richard Morris · Fabio Ramos · Peter Dayan · Bernard Balleine -
2017 : 6 Spotlight Talks (3 min each) »
Mennatullah Siam · Mohit Prabhushankar · Priyam Parashar · Mustafa Mukadam · hengshuai yao · Ransalu Senanayake -
2016 Poster: Spatio-Temporal Hilbert Maps for Continuous Occupancy Representation in Dynamic Environments »
Ransalu Senanayake · Lionel Ott · Simon O'Callaghan · Fabio Ramos -
2016 Poster: Kernel Observers: Systems-Theoretic Modeling and Inference of Spatiotemporally Evolving Processes »
Hassan A Kingravi · Harshal R Maske · Girish Chowdhary -
2014 Poster: On Integrated Clustering and Outlier Detection »
Lionel Ott · Linsey Pang · Fabio Ramos · Sanjay Chawla