This website was used for the 2017 instance of this workshop.
Please visit ml4physicalsciences.github.io for up-to-date information.

About

Physical sciences span problems and challenges at all scales in the universe: from finding exoplanets and asteroids in trillions of sky-survey pixels, to automatic tracking of extreme weather phenomena in climate datasets, to detecting anomalies in event streams from the Large Hadron Collider at CERN. Tackling a number of associated data-intensive tasks, including, but not limited to, regression, classification, clustering, dimensionality reduction, likelihood-free inference, generative models, and experimental design are critical for furthering scientific discovery. The Deep Learning for Physical Sciences (DLPS) workshop invites researchers to contribute papers that demonstrate progress in the application of machine and deep learning techniques to real-world problems in physical sciences (including the fields and subfields of astronomy, chemistry, Earth science, and physics).

We will discuss research questions, practical implementation challenges, performance / scaling, and unique aspects of processing and analyzing scientific datasets. The target audience comprises members of the machine learning community who are interested in scientific applications and researchers in the physical sciences. By bringing together these two communities, we expect to strengthen dialogue, introduce exciting new open problems to the wider NeurIPS community, and stimulate production of new approaches to solving science problems. Invited talks from leading individuals from both communities will cover the state-of-the-art techniques and set the stage for this workshop.

NeurIPS 2017

The Deep Learning for Physical Sciences (DLPS) 2017 workshop will be held on December 8, 2017 as a part of the 31st Annual Conference on Neural Information Processing Systems, at the Long Beach Convention & Entertainment Center, Long Beach, CA, United States. Please check the main conference website for information about registration, schedule, venue, and travel arrangements.

Photos from 2017 workshop

Survey

Help us improve the workshop by filling out this quick post-event survey.

Your opinion matters!

Schedule

Invited speakers

Schedule

08:50 – 09:00 Introduction

Atılım Güneş Baydin (University of Oxford)
Introduction and opening remarks [slides]

Morning session chair:
Michela Paganini (Yale University)
09:00 – 09:40 Invited talk 1

Max Welling (University of Amsterdam)
Deep recurrent inverse modeling for radio astronomy and fast MRI imaging [slides]

09:40 – 10:00 Contributed talk 1

Isaac Henrion (New York University)
Neural Message Passing for Jet Physics [slides]

10:00 – 10:20 Contributed talk 2

Auralee Edelen (Colorado State University and Fermilab)
Using Neural Network Control Policies For Rapid Switching Between Beam Parameters in a Free Electron Laser [slides]

10:20 – 11:00 Poster session and coffee break
11:00 – 11:40 Invited talk 2

Gilles Louppe (University of Liège)
Adversarial Games for Particle Physics [slides]

11:40 – 12:00 Contributed talk 3

Dustin Tran (Columbia University)
Implicit Causal Models for Genome-wide Association Studies [slides]

12:00 – 12:20 Contributed talk 4

Aditya Grover (Stanford University)
Graphite: Iterative Generative Modeling of Graphs [slides]

12:20 – 12:25 Sponsor presentation

Hanlin Tang (Intel Nervana)
Intel AI Lab
[slides]
12:25 – 14:00 Lunch break
Afternoon session chair:
Savannah Thais (Yale University)
14:00 – 14:40 Invited talk 3

Iain Murray (University of Edinburgh)
Learning priors, likelihoods, or posteriors [slides]

14:40 – 15:00 Contributed talk 5

Daniel George (University of Illinois at Urbana-Champaign)
Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation with Real LIGO Data [slides]

15:00 – 16:00 Poster session and coffee break
16:00 – 16:40 Invited talk 4

Juan Carrasquilla (D-Wave Systems / Vector Institute for Artificial Intelligence)
A machine learning perspective on the many-body problem in classical and quantum physics [slides]

16:40 – 17:20 Invited talk 5

Anatole von Lilienfeld (University of Basel)
Quantum Machine Learning [slides]

17:20 – 17:40 Contributed talk 6

Anuj Karpatne (University of Minnesota)
How Can Physics Inform Deep Learning Methods in Scientific Problems?: Recent Progress and Future Prospects [slides]

17:40 – 18:40 Panel session

Moderator:
Kyle Cranmer (New York University)

Panelists:
Iain Murray (University of Edinburgh)
Max Welling (University of Amsterdam)
Juan Carrasquilla (D-Wave Systems / Vector Institute for Artificial Intelligence)
Gilles Louppe (University of Liège)
George Dahl (Google Brain)
Anatole von Lilienfeld (University of Basel)

18:40 – 18:45 Closing remarks

Accepted papers

Adversarial learning to eliminate systematic errors: a case study in High Energy Physics [pdf]
Victor Estrade, Cecile Germain, Isabelle Guyon and David Rousseau
Variational Inference over Non-differentiable Cardiac Simulators using Bayesian Optimization [pdf]
Adam McCarthy, Blanca Rodriguez and Ana Minchole
Deep topology classifiers for a more efficient trigger selection at the LHC [pdf]
Daniel Weitekamp III, Thong Q. Nguyen, Dustin Anderson, Roberto Castello, Maurizio Pierini, Maria Spiropulu and Jean-Roch Vlimant
Nanophotonic Particle Simulation and Inverse Design Using Artificial Neural Networks [pdf]
John Peurifoy, Yichen Shen, Li Jing, Yi Yang, Fidel Cano-Renteria, Brendan Delacy, Max Tegmark, John Joannopoulos and Marin Soljacic
FlareNet: A Deep Learning Framework for Solar Phenomena Prediction [pdf]
Sean McGregor, Dattaraj Dhuri, Anamaria Berea and Andrés Muñoz-Jaramillo
Solving differential equations with unknown constitutive relations as recurrent neural networks [pdf]
Tobias Hagge, Panos Stinis, Enoch Yeung and Alexandre Tartakovsky
Convolutional Neural Networks for Electron Neutrino and Electron Shower Energy Reconstruction in the NOvA Detectors [pdf]
Lars Hertel, Lingge Li, Pierre Baldi and Jianming Bian
Deep Learning Reconstruction of Ultra-Short Pulses [pdf]
Tom Zahavy, Alex Dikopoltsev, Shie Mannor, Oren Cohen and Moti Segev
Towards a Hybrid Approach to Physical Process Modeling [pdf]
Emmanuel de Bézenac, Arthur Pajot and Patrick Gallinari
DeepJet: Generic physics object based jet multiclass classification for LHC experiments [pdf]
Markus Stoye, Jan Kieseler, Mauro Verzetti, Huilin Qu, Loukas Gouskos, Anna Stakia and CMS Collaboration
Graph Memory Networks for Molecular Activity Prediction [pdf]
Trang Pham, Truyen Tran and Svetha Venkatesh
PSIque: Next Sequence Prediction of Satellite Images using a Convolutional Sequence-to-Sequence Network [pdf]
Seungkyun Hong, Seongchan Kim and Sa-Kwang Song
ConvWave: Searching for Gravitational Waves with Fully Convolutional Neural Nets [pdf]
Timothy Gebhard, Niki Kilbertus, Giambattista Parascandolo, Ian Harry and Bernhard Schölkopf
Implicit Causal Models for Genome-wide Association Studies [pdf]
Dustin Tran and David Blei
Calorimetry with Deep Learning: Particle Classification, Energy Regression, and Simulation for High-Energy Physics [pdf]
Benjamin Hooberman, Amir Farbin, Gulrukh Khattak, Vitória Pacela, Maurizio Pierini, Jean-Roch Vlimant, Maria Spiropulu, Wei Wei, Matt Zhang and Sofia Vallecorsa
Using Neural Network Control Policies For Rapid Switching Between Beam Parameters in a Free Electron Laser [pdf]
Auralee Edelen, Jonathan Edelen, Sandra Biedron, Stephen Milton and Peter van der Slot
Data Quality Network for Spatiotemporal Forecasting [pdf]
Sungyong Seo, Arash Mohegh, George Ban-Weiss and Yan Liu
Deep Learning for Real-time Gravitational Wave Detection and Parameter Estimation with Real LIGO Data [pdf]
Daniel George and E. A. Huerta
How Can Physics Inform Deep Learning Methods in Scientific Problems?: Recent Progress and Future Prospects [pdf]
Anuj Karpatne, William Watkins, Jordan Read and Vipin Kumar
Segmenting and Tracking Extreme Climate Events using Neural Networks [pdf]
Mayur Mudigonda, Soo Kim, Ankur Mahesh, Samira Kahou, Karthik Kashinath, Dean Williams, Vincent Michalski, Travis O'Brien and Mr Prabhat
Graphite: Iterative Generative Modeling of Graphs [pdf]
Aditya Grover, Aaron Zweig and Stefano Ermon
Searching for Exoplanets Using Artificial Intelligence [pdf]
Kyle A. Pearson, Leon Palafox and Caitlin A. Griffith
Searching for Long-Period Comets with Deep Learning Tools [pdf]
Susana Zoghbi, Marcelo De Cicco, Antonio Ordonez, Andres Plata Stapper, Jack Collison, Peter Gural, Siddha Ganju, Jose Luis Galache and Peter Jenniskens
Survey of Machine Learning Techniques for High Energy Electromagnetic Shower Classification [pdf]
Michela Paganini, Luke de Oliveira and Benjamin Nachman
Glitch Classification and Clustering for LIGO with Deep Transfer Learning [pdf]
Daniel George, Hongyu Shen and E. A. Huerta
Tips and Tricks for Training GANs with Physics Constraints [pdf]
Luke de Oliveira, Michela Paganini and Benjamin Nachman
Towards understanding feedback from supermassive black holes using convolutional neural networks [pdf]
Stanislav Fort
Particle Track Reconstruction with Deep Learning [pdf]
Steven Farrell, Paolo Calafiura, Mayur Mudigonda, Mr. Prabhat, Dustin Anderson, Josh Bendavid, Maria Spiropoulou, Jean-Roch Vlimant, Stephan Zheng, Giuseppe Cerati, Lindsey Gray, Jim Kowalkowski, Panagiotis Spentzouris, Aristeidis Tsaris and Daniel Zurawski
Neural Message Passing for Jet Physics [pdf]
Isaac Henrion, Kyle Cranmer, Joan Bruna, Kyunghyun Cho, Johann Brehmer, Gilles Louppe and Gaspar Rochette
Improvements to Inference Compilation for Probabilistic Programming in Large-Scale Scientific Simulators [pdf]
Mario Lezcano Casado, Atılım Güneş Baydin, David Martinez Rubio, Tuan Anh Le, Frank Wood, Lukas Heinrich, Gilles Louppe, Kyle Cranmer, Wahid Bhimji, Karen Ng and Prabhat

Call for papers

We invite researchers to submit papers particularly in the following and related areas:

  • Application of machine and deep learning to physical sciences
  • Generative models
  • Likelihood-free inference
  • Variational inference
  • Simulation-based models
  • Implicit models
  • Probabilistic models
  • Model interpretability
  • Approximate Bayesian computation
  • Strategies for incorporating prior scientific knowledge into machine learning algorithms
  • Experimental design
  • Any other area related to the subject of the workshop

Submissions of completed projects as well as high-quality works in progress are welcome. All accepted papers will be made available on the workshop website and presented as posters or contributed talks during the workshop. As this does not constitute an archival publication or formal proceedings, authors are free to publish their extended work elsewhere. Submissions will be kept confidential until they are accepted and authors confirm that they can be included in the workshop. If a submission is not accepted, or withdrawn for any reason, it will be kept confidential and not made public.

Up to 6 accepted submissions will be selected for 20-minute contributed talks.

Submission instructions

Submissions should be anonymized short papers up to 4 pages in PDF format, typeset using the NeurIPS style. References do not count towards the page limit. Appendices are discouraged, and any appendix pages will be included in the 4-page limit. A workshop-specific modified NeurIPS style file will be provided for the camera-ready versions, after the author notification date.

Submissions are handled through the EasyChair website for DLPS 2017. Please note that at least one coauthor of each accepted paper will be expected to attend the workshop in person to present a poster or give a contributed talk.

Submit paper

Travel support and complimentary registration

We have a number of complimentary workshop registrations that will be handed out to paper authors. We also have travel support available particularly for students and junior researchers, thanks to the generous support from our sponsors.

To apply for complimentary registration and travel support, please submit your paper and get in touch with the contact address listed at the bottom of this page, briefly describing your circumstances (e.g., undergraduate or graduate student, postdoctoral researcher, junior faculty, etc.) and intended travel plans (travel origin).

Important dates

  • Submission deadline: November 1, 2017 November 3, 2017, 23:59 PDT
  • Author notification: November 10, 2017 November 13, 2017
  • NeurIPS deadline to cancel registration: November 16, 2017
  • Camera-ready (final) paper deadline: December 1, 2017
  • Workshop: December 8, 2017

Information for Accepted Papers and Posters

Please produce a "camera-ready" (final) version of your accepted paper by replacing the "nips_2017.sty" style file with the "nips_dlps_2017.sty" file available here and using the "final" option (that is, "\usepackage[final]{nips_dlps_2017}") to include author and affiliation information. The modified style file replaces the first page footer to refer to the workshop instead of the main conference. It is acceptable if your paper goes above the 4-page limit (excluding references) due to author and affiliation information appearing. It is not acceptable to make revisions beyond minor corrections and to include material that was not present in the reviewed version of your paper.

Please upload the final PDF as an updated version in your existing submission on EasyChair. The uploaded final version will be hosted in the workshop website in an "accepted papers" section.

For your posters, we suggest A0 size measuring 841 × 1189 mm (33.1 × 46.8 in). Note that the workshop venue cannot accommodate posters larger than 910 × 1220 mm (36 × 48 in). All accepted papers that do not have an oral presentation (contributed talks in the schedule) are expected to be presented as posters. Posters are optional for oral presenters.

Organizers

Location

Promenade Room 104C, Long Beach Convention & Entertainment Center, 300 E Ocean Blvd, Long Beach, CA 90802, United States