Conformal Prediction in 2022
Conformal inference methods are becoming all the rage in academia and industry alike. In a nutshell, these methods deliver exact prediction intervals for future observations without making any distributional assumption whatsoever other than having iid, and more generally, exchangeable data. This talk will review the basic principles underlying conformal inference and survey some major contributions that have occurred in the last 2-3 years or. We will discuss enhanced conformity scores applicable to quantitative as well as categorical labels. We will also survey novel methods which deal with situations, where the distribution of observations can shift drastically — think of finance or economics where market behavior can change over time in response to new legislation or major world events, or public health where changes occur because of geography and/or policies. All along, we shall illustrate the methods with examples including the prediction of election results or COVID19-case trajectories.
NeurIPS 2022 Competition Track: Overview & Results
An introduction by Isabelle Guyon and Evelyne Viegas followed by posters presentations of competition results with the main organizers.
ML Safety NeurIPS Social
As ML systems increase in capabilities and integration in society, their safety and alignment grow in importance. We welcome participants interested in or working on AI safety topics for a semi-structured social meetup at NeurIPS 2022. Meet peers who may have read your papers or authors you've read, and build new friendships and collaboration opportunities. We recommend using Swapcard, to request one-one-one meetings with other attendants. Example topics include "Establishing trust in advanced AI systems," "Concrete Threat Models," and "AI Governance and Policy", "Model interpretability", ...
NeurIPS 2022 Space & ML Social Event
The NeurIPS 2022 Space & ML social will kick off with an opening lighting round of callbacks, where guests can make their brief case for the potential of ML on (but not limited to) areas such as the following: climate change (predicting wildfires, preventing floods), space exploration, early event detection and situational awareness, fairness and equity in algorithmic decision-making, or open-source science. Following this lightning round, we will organize a social experience for attendees to meet each other and foster co-opetition (cooperative competition) in a space pub quiz, with special guests from our space industry partners (NASA, US Department of Energy, ESA, Australia Space Agency, etc).
RL Social
The purpose of this social is to provide an opportunity for the researchers in the RL community, both young and senior, and from both industry and academia to interact, meet each other and exchange ideas. In particular, we the organizers will invite RL experts from the many sub-areas of RL to attend and provide younger researchers with the opportunity to meet them and discuss any topics which might be of interest in their research and career.
Ethics Review - Open Discussion
Ethical review has become an intrinsic part of the NeurIPS review process, but many unanswered question remain around topics such as: the role of ethics review; current and potential future ethical issues (consent, privacy, copyright); ethics awareness/visibility (e.g. how to help technical reviewers with access to resources to adequately flag ethical issues in papers), etc. We will start the townhall with a brief panel discussions and then open up the floor to community input/feedback and answer any questions that audience members will have.
| NeurIPS uses cookies for essential functions only. We do not sell your personal information. Our Privacy Policy » |