Searching for New Physics with Reinforcement Learning
A reinforcement-learning method is being used to search the full SMEFT operator space for explanations of particle-physics anomalies.
The paper frames the task as a way to reduce human bias in choosing which operators might matter. It tests the method on the CDF W-mass anomaly, where it says the system reproduces and improves on known results. The authors also apply it to a harder multi-anomaly case and report that it still finds operators that explain the data. ArXiv · AI/CL/LG's note
The paper frames the task as a way to reduce human bias in choosing which operators might matter. It tests the method on the CDF W-mass anomaly, where it says the system reproduces and improves on known results. The authors also apply it to a harder multi-anomaly case and report that it still finds operators that explain the data. ArXiv · AI/CL/LG's note
score 4