Workshop description

Call for abstract:

We welcome 2-page extended abstracts on topics relating to challenges in forecasting, machine learning for spatio-temporal forecasting and related application areas. Selected papers will be presented primarily as posters. Exceptional contributions may be given oral presentations. Abstract should be submitted by Nov 14th, 2016 by sending email to

We plan to invite the authors of high-quality submissions to submit extended versions of their work for another round of reviews and publication in post-workshop proceedings. For submission details such as TeX template instructions please inquire at the email address given above.


Hilton Diag. Mar, Blrm. B, Sat Dec 10, 09:00 AM

Morning session High-impact applications in science and industry: what is forecasting good for?
9:00 - 9:30 Welcome and introduction to spatiotemporal forecasting: platforms, tools, datasets and challenges. Florin Popescu (Fraunhofer Institute, Germany).
9:30 - 10:00 Invited talk: Forecasting for Electrical Transmission Grid: Antoine Marot (RTE: Réseau de transport d’électricité, FR).
10:00- 10:30 Invited talk: Danny Silver (Acadia University, CA): Forecasting using machine learning techniques in energy and agriculture.
11:00- 12:00 Open discussion: Application areas of advanced forecasting methods.
12:00- 14:30 Lunch, poster session.
14:30- 15:00 Invited talk: Alexander Statnikov (American Express): Financial Risk Forecasting.
15:00-15:30 Coffee Break
15:30- 16:30 Open/guided discussion: Platforms for forecasting.
Special session: Learning theory and practice: what is good forecasting?
16:30- 17:00 Invited talk: Sven Crone (Lancaster Centre for Forecasting, Lancaster Univ.): From Hype-Cycle to Reality of Predictive Analytics (a Time Series Forecasting perspective)
17:00- 17:45 Invited talk: Spatiotemporal online learning with expert advice, with applications to climate science and finance: Claire Monteleoni (George Washington University)
17:45- 19:00 Final discussion (moderated by Program Committee)

Click here to download the full schedule in a PDF format.

Workshop organizers make last-minute changes to their schedule. Download this document again to get the lastest changes, or use the NIPS mobile application.

Topics of interest


  • Novel methods related to spatio-temporal forecasting: image parametrization and analysis such as dimensionality reduction, time series prediction methods over large state spaces, feature selection for forecasting, causal feature selection in time series.
  • Novel methods to deal with wide data bandwidths in forecasting.
  • Novel methods for adaptive, quick turnaround deep learning in forecasting.
  • Benchmarks for forecasting video sequences
  • Benchmarks for forecasting of time series
  • Incremental learning and flexible resource-constrained learning.
  • Automatic machine learning for forecasting.
  • Spatio-temporal datasets for forecasting in agriculture, epidemiology, geosciences, economics, animate motion


  • Statistical comparison of quality in forecasting systems
  • Stochastic forecasting evaluation
  • Combinations of loss types for increased forecasting reliability
  • Robust tail-estimation and distribution representation of stochastic forecasts


  • Computational languages, platforms and toolboxes for forecasting
  • Adaptation of computer vision toolboxes for video understanding and forecasting
  • Big open data pipelines for forecasting

Applications: including:

  • Energy sector, grid, renewables
  • Video-based applications, compression, predictive control systems using video streams
  • Stochastic predictions in geosciences and meteorology using ML
  • Epidemiology
  • Economics
  • Agriculture
  • Multi-disciplinary and mult-modal convergence for event and trajectory forecasting.


  • Florin Popescu -
  • Sergio Escalera -
  • Xavier Baró -
  • Stephane Ayache -
  • Isabelle Guyon -