About this Research Topic
After having analyzed ten years of proton-proton collision data, no direct evidence of new particles being produced at the Large Hadron Collider has been observed.
Making sure that the largest amount of useful data is recorded, as well as developing novel methods for detecting increasingly rare events, will therefore be highly important in the search for New Physics during the next years of LHC data taking.
Effort has been made by all LHC collaborations in optimizing and improving the way in which collider data is selected, processed, and analyzed in particle physics experiments. This can be done for instance by refining or changing the way events are selected in detector data acquisition systems, or by using non-standard data analysis workflows. These efforts allow us to probe new, unexplored regions in parameter space, as well as to obtain levels of precision for measurements that have up until now been out of reach for experiments.
One promising technique for finding signals in large datasets involves the use of unsupervised or semi-supervised Machine Learning (ML) techniques for anomaly detection in particle physics, like autoencoders or other density models. By reformulating searches for new phenomena as out-of-distribution detection tasks, one can take advantage of the large amount of unlabeled data in particle physics experiments and look for new physics in a model-agnostic way.
Some examples of methods for improving the precision of our measurements, as well as for searching for New Physics signals in particle physics include:
- Improving algorithms for data acquisition and event selection (triggering) in detector systems
- analysing high-rate datasets of collider events where a significant part of the initial data processing is done on timescales of milliseconds (real-time analysis) within the experiment’s initial event selection (trigger) systems
- searching for new physics signals by looking for anomalous data in a background of Standard Model interactions, and,
- dimensionality reduction and clustering of physics data.
In this Research Topic, we are interested in new algorithms, techniques, and case studies that attempt to improve or change the way in which data is selected and analyzed in high energy physics experiments. This includes novel event selection algorithms, data collection techniques, or new analysis strategies.
Making sure that the largest amount of useful data is recorded, as well as developing novel methods for detecting increasingly rare events, will therefore be highly important in the search for New Physics during the next years of LHC data taking.
Effort has been made by all LHC collaborations in optimizing and improving the way in which collider data is selected, processed, and analyzed in particle physics experiments. This can be done for instance by refining or changing the way events are selected in detector data acquisition systems, or by using non-standard data analysis workflows. These efforts allow us to probe new, unexplored regions in parameter space, as well as to obtain levels of precision for measurements that have up until now been out of reach for experiments.
One promising technique for finding signals in large datasets involves the use of unsupervised or semi-supervised Machine Learning (ML) techniques for anomaly detection in particle physics, like autoencoders or other density models. By reformulating searches for new phenomena as out-of-distribution detection tasks, one can take advantage of the large amount of unlabeled data in particle physics experiments and look for new physics in a model-agnostic way.
Some examples of methods for improving the precision of our measurements, as well as for searching for New Physics signals in particle physics include:
- Improving algorithms for data acquisition and event selection (triggering) in detector systems
- analysing high-rate datasets of collider events where a significant part of the initial data processing is done on timescales of milliseconds (real-time analysis) within the experiment’s initial event selection (trigger) systems
- searching for new physics signals by looking for anomalous data in a background of Standard Model interactions, and,
- dimensionality reduction and clustering of physics data.
In this Research Topic, we are interested in new algorithms, techniques, and case studies that attempt to improve or change the way in which data is selected and analyzed in high energy physics experiments. This includes novel event selection algorithms, data collection techniques, or new analysis strategies.
Keywords: Trigger, Scouting, Anomaly Detection, turbo, trigger-level analysis, real-time analysis, outlier detection
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
