ISNET-11: Information and Statistics in Nuclear Experiment and Theory

***WARNING: for hotel bookings do not give your data to any external services upon email request; there are known cases of fraud***

This workshop will gather leading international researchers in nuclear physics, statistics, and applied mathematics to discuss how statistical methods can enable progress on the frontiers of nuclear physics, spawn new directions, and catalyze techniques that ensure best use of data taken in nuclear-physics experiments.

Uncertainty quantification is now a key component of nuclear physics research. Statistical inference methods form an increasingly integral part of theoretical, experimental, and computational physics. Rapid advances in computational statistics and machine learning enable exciting possibilities: combining model forecasts, exploring the intricate structure and limits of theoretical models, and guiding future experiments to the most information-rich domains. These techniques can also spur greater computational efficiency, an important consideration given the computer-hungry nature of state-of-the-art nuclear-physics calculations.

We envision a meeting addressing the following topics

1. Intrusive/non-intrusive emulation strategies and model-order reduction
2. New approaches to model calibration
3. Multi-model forecasting: model selection, averaging, and mixing
4. Using existing data & planning for future data
5. Putting it all together

Each topic will involve at least one longer talk by a statistician or applied mathematician that addresses methodological questions regarding scientific inference. Each day will include multiple talks on nuclear-physics applications pertinent to the topic for that day. This format ensures that nuclear physicists from different sub-fields can participate vigorously throughout the workshop. We will also organize one longer discussion session per day, led by one-two experienced discussion leaders who will be responsible for highlighting and expanding on selected questions.

For some participants it will be their first time attending an ISNET meeting. To enable researchers to enter the field and pick up statistical methods in their research we will organize an hour-long introduction, tentatively called “No stupid questions” , at the beginning of the workshop—an approach we have implemented successfully in previous ISNET meetings. If desired by participants we will have subsequent “no stupid questions” sessions before the formal beginning of the workshop each day.