Nanosecond anomaly detection with decision trees and real-time application to exotic Higgs decays
Authors:
Stephen Roche,
Quincy Bayer,
Benjamin Carlson,
William Ouligian,
Pavel Serhiayenka,
Joerg Stelzer,
Tae Min Hong
Abstract:
We present an interpretable implementation of the autoencoding algorithm, used as an anomaly detector, built with a forest of deep decision trees on FPGA, field programmable gate arrays. Scenarios at the Large Hadron Collider at CERN are considered, for which the autoencoder is trained using known physical processes of the Standard Model. The design is then deployed in real-time trigger systems fo…
▽ More
We present an interpretable implementation of the autoencoding algorithm, used as an anomaly detector, built with a forest of deep decision trees on FPGA, field programmable gate arrays. Scenarios at the Large Hadron Collider at CERN are considered, for which the autoencoder is trained using known physical processes of the Standard Model. The design is then deployed in real-time trigger systems for anomaly detection of unknown physical processes, such as the detection of rare exotic decays of the Higgs boson. The inference is made with a latency value of 30 ns at percent-level resource usage using the Xilinx Virtex UltraScale+ VU9P FPGA. Our method offers anomaly detection at low latency values for edge AI users with resource constraints.
△ Less
Submitted 15 April, 2024; v1 submitted 7 April, 2023;
originally announced April 2023.
Nanosecond machine learning regression with deep boosted decision trees in FPGA for high energy physics
Authors:
Benjamin Carlson,
Quincy Bayer,
Tae Min Hong,
Stephen Roche
Abstract:
We present a novel application of the machine learning / artificial intelligence method called boosted decision trees to estimate physical quantities on field programmable gate arrays (FPGA). The software package fwXmachina features a new architecture called parallel decision paths that allows for deep decision trees with arbitrary number of input variables. It also features a new optimization sch…
▽ More
We present a novel application of the machine learning / artificial intelligence method called boosted decision trees to estimate physical quantities on field programmable gate arrays (FPGA). The software package fwXmachina features a new architecture called parallel decision paths that allows for deep decision trees with arbitrary number of input variables. It also features a new optimization scheme to use different numbers of bits for each input variable, which produces optimal physics results and ultraefficient FPGA resource utilization. Problems in high energy physics of proton collisions at the Large Hadron Collider (LHC) are considered. Estimation of missing transverse momentum (ETmiss) at the first level trigger system at the High Luminosity LHC (HL-LHC) experiments, with a simplified detector modeled by Delphes, is used to benchmark and characterize the firmware performance. The firmware implementation with a maximum depth of up to 10 using eight input variables of 16-bit precision gives a latency value of O(10) ns, independent of the clock speed, and O(0.1)% of the available FPGA resources without using digital signal processors.
△ Less
Submitted 12 July, 2022;
originally announced July 2022.
