Searcharxiv⌕ Search

arXiv subjects

Marco Coïsson

Publications and source records attributed to Marco Coïsson.

2 recordsLinked to original sources

Beauty and charm jet identification with end-to-end training on a superconducting quantum processor

The identification of heavy-flavor jets, i.e. collimated streams of particles originating from the fragmentation of beauty and charm quarks produced in proton-proton collisions, represents one of the most challenging classification tasks in measurements at Large Hadron Collider experiments. In recent years, advanced machine learning techniques have been successfully employed to address this problem, while quantum machine learning has emerged as a promising alternative approach. However, the high dimensionality of the problem, typically involving hundreds of observables, raises an open question: whether the current generation of quantum computers, characterized by limited qubit count, can effectively tackle such complex tasks. In this work, a variational quantum circuit for the classification of b- and c-jets at the LHCb experiment is implemented on a 5-qubit superconducting quantum computer manufactured by IQM and installed at the Politecnico di Torino. To the best of our knowledge, this is the first demonstration of a quantum algorithm for jet identification using simulated data from an LHC experiment in which all stages of the machine-learning workflow, including training and testing, are performed on real quantum hardware. The results demonstrate that the impact of hardware-induced quantum noise does not significantly degrade the performance. A framework for the evaluation of the training time on the quantum hardware and for the circuit optimization is also proposed. These findings pave the way for future applications on larger quantum devices based on superconducting qubits, featuring a larger number of qubits enabling the encoding of information beyond the reach of classical simulation and providing a step toward investigating whether quantum advantage may eventually be achievable in high-energy physics classification tasks.

quant-ph↗

Multi-analytical characterization of Fe-rich magnetic inclusions in diamonds

Magnetic mineral inclusions, as iron oxides or sulfides, occur quite rarely in natural diamonds. Nonetheless, they represent a key tool not only to unveil the conditions of formation of host diamonds, but also to get hints about the paleointensity of the geomagnetic field present at times of the Earth's history otherwise not accessible. This possibility is related to their capability to carry a remanent magnetization dependent on their magnetic history. However, comprehensive experimental studies on magnetic inclusions in diamonds have been rarely reported so far. Here we exploit X-ray diffraction, Synchrotron-based X-ray Tomographic Microscopy and Alternating Field Magnetometry to determine the crystallographic, morphological and magnetic properties of ferrimagnetic Fe-oxides entrapped in diamonds coming from Akwatia (Ghana). We exploit the methodology to estimate the natural remanence of the inclusions, associated to the Earth's magnetic field they experienced, and to get insights on the relative time of formation between host and inclusion systems. Furthermore, from the hysteresis loops and First Order Reversal Curves we determine qualitatively the anisotropy, size and domain state configuration of the magnetic grains constituting the inclusions.

physics.geo-ph↗