Abstract Understanding the evolution of a process over space and time is fundamental to a variety of disciplines. Such phenomena that exhibit dynamics in both space and time include propagation of diseases, evolution of agro-ecosystems, 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 becoming crucial for an agent to successfully interact with 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 control; and (3) Encouraging conversation between theoreticians and practitioners to develop robust predictive models. Keywords Theory: stochastic processes, deep learning/convolutional LSTM, control theory, Bayesian filtering, kernel methods, time-frequency analysis, 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 sciences and economics: predictive policing, population mapping, poverty mapping, food resources, agricultural monitoring and control, etc. Engineering/robotics: active data collection, traffic modeling, motion prediction, fluid dynamics, music representation and generation, analysis of video data, multi-sensor fusion, etc. Updates * We have 20 tickets from the reserved tickets pool available for speakers of the workshop. Important Dates Camera ready deadline: November 30, 2018 11:59 pm AOE Workshop: December 07, 2018 8.00-6:30 (Room 513ABC, Palais des Congrès de Montréal, Montréal, Canada) Camera-ready submission instructions 1. Replace the original NIPS latex style file with this modified latex style file. The only difference in the new style file is that the details of the workshop has been added to the footnote. 2. Change \usepackage{nips_2018} to \usepackage[final]{nips_2018} in the tex file. Make sure all author details are added. 3. Please note that the maximum number of pages for the main content of the paper is four, excluding references and supplementary materials. If the paper accompanies supplementary files (pdfs,
videos, etc.), please upload them to your own file hosting service
(dropbox, youtube, github, etc.) and clearly provide the links just below the
abstract. Alternatively, you might append the paper (after references) with supplementary materials. 4. Replace the existing pdf file in the OpenReview System with the camera-ready pdf. Attending NIPS
|