Speaker
Description
Anomaly detection (AD) has recently emerged as an exciting alternative to conventional search strategies in high energy physics. The integration of these techniques into trigger systems is even more recent, but represents a crucial step in expanding the coverage of LHC triggers. In this paper, we explore the direct comparison, as well as combination, of two compression techniques for variational autoencoder (VAE) AD trigger algorithms: (1) utilizing only latent-space derived variables and therefore requiring only half of the VAE (to chop) and (2) applying knowledge distillation (KD) to distill the VAE into a student architecture (not to chop). We demonstrate the feasibility of deploying both techniques on an FPGA within the resource and latency constraints of the LHC trigger environment and further find that a combination of the two leads to the smallest models that maintain, and in some cases, improve, performance with respect to the original VAE architecture.
| Do you plan to submit a 4-page extended abstract on OpenReview (only for Presentations/Posters)? | No |
|---|