NeurIPS workshop on
Interpretable Inductive Biases
and Physically Structured Learning

December 11th–12th, 2020


Over the last decade, deep networks have propelled machine learning to accomplish tasks previously considered far out of reach, human-level performance in image classification and game-playing. However, research has also shown that the deep networks are often brittle to distributional shifts in data: it has been shown that human-imperceptible changes can lead to absurd predictions. In many application areas, including physics, robotics, social sciences and life sciences, this motivates the need for robustness and interpretability, so that models can be trusted in practical applications. Interpretable and robust models can be constructed by incorporating prior knowledge within the model or learning process as an inductive bias, thereby avoiding overfitting and making the model easier to understand for scientists and non-machine-learning experts. In this workshop, we bring together researchers from different application areas to study the inductive biases that can be used to obtain interpretable models. We invite speakers from physics, robotics, and other related areas to share their ideas and success stories on inductive biases. We also invite researchers to submit extended abstracts for contributed talks and posters to initiate lively discussion on inductive biases and foster this growing community of researchers.

Call for Papers

We invite researchers to submit work in any of the following areas.

  • Inductive biases for machine learning.
  • Interpretable deep networks.
  • Encoding prior information into machine learning models.
  • Connections between machine learning and differential equations.
  • Connections between machine learning, geometry, and mechanics.
  • Applications to physics, robotics, biology, chemistry, and climate science.

A submission should be made in the form of an short 4-page paper using the NeurIPS style. Manuscripts should be anonymized in accordance with the same rules as NeurIPS papers. References can extend as far beyond the 4-page limit as needed: we encourage use of BibLaTeX, which can be used with the NeurIPS style via the nonatbib package option in the latter. If the submitted work has previously appeared in a journal or refereed workshop or conference proceedings, including the current NeurIPS proceedings to be published, the workshop submission should extend the previous work.

Submit your workshop paper by October 9th via CMT

Key Dates

Date Event
Oct 09 2020 Workshop Paper Submission Deadline
Oct 23 2020 Workshop Paper Acceptance Notification
Nov 14 2020 Talk Recording Deadline
Dec 11/12 2020 Workshop Day