ICLR-21 Workshop on Responsible AI
An interdisciplinary perspective on practical limitations and tradeoffs among fairness, explainability, safety, robustness, and beyond.
Call for Papers
The workshop solicits novel contributions that touch on any of the topics that relate broadly to responsible AI, including but not limited to fairness, explainability, safety, and robustness. We especially welcome submissions that address the limitations of recent computational approaches to fairness, explainability, robustness and safety in ML, and advance the community’s understanding of the tradeoffs among these concepts.
We welcome submissions in the following areas:
Limitations and tradeoffs among responsible ML objectives: understanding the limitations of existing computational approaches to fairness, explainability, robustness, and safety, and the conditions under which they compete against each other as well as accuracy.
Long-term impact disparities and unintended consequences of responsible ML: Capturing and mitigating the long-term risks and unintended consequences of responsible ML methods.
Algorithmic and learning approaches to responsible ML: Where do we stand in evaluating and incorporating fairness, safety, interpretability, and robustness constraints into our learning pipelines (e.g., in deep learning pipelines)?
Sources and causes of failure vis-a-vis responsible ML desiderata: including but not limited to the training data, regularization, choice of objective functions, testing and deployment conditions, and beyond.
Incorporating domain knowledge and insights from social sciences: practical incorporation of insights from social sciences and domain expertise into the design, deployment, and maintenance of ML systems.
Responsible ML in application: societal and ethical considerations in applying ML to criminal justice, health care and medicine, poverty and homelessness, and education.
Important Dates
Paper Submission Deadline: February 26, 2021
Decision Notification: March 26, 2021
Camera Ready Paper Deadline: April 15, 2021
Workshop Date: May 8, 2021
Submission Instructions
Submitted papers can be up to 4 pages in ICLR submission format (double-blind), excluding references and appendix. We allow an unlimited number of pages for references and supplementary material, but reviewers are not required to review the supplementary material. Accepted papers will be presented at the workshop as contributed talks and/or posters. At the discretion of authors, accepted papers can be published through the workshop website.
By submitting their work to the workshop, authors agree to serve as reviewers.