Classical vs quantum: Comparing tensor network based quantum circuits on Large Hadron Collider data
This research paper explores the use of Tensor Networks (TNs) for machine learning tasks, specifically focusing on the discrimination of top quarks from QCD background in simulated Large Hadron Collider (LHC) data. The study compares the performance of classical TNs with their quantum counterparts, known as Quantum Tensor Networks (QTNs), across different architectures such as Matrix Product States (MPS), Tree Tensor Networks (TTN), and Multi-scale Entanglement Renormalisation Ansatz (MERA). The paper finds that QTNs achieve higher accuracy with significantly fewer trainable parameters than classical TNs, highlighting the potential of QTNs for machine learning applications in high-energy physics. To address the limitations of QTNs due to the small number of qubits available in current quantum devices, the paper proposes hybrid classical-quantum TN architectures, where a classical TN layer processes the data before being inputted into a QTN for classification. The study concludes that hybrid architectures offer a promising approach for leveraging the benefits of both classical and quantum computation in machine learning tasks, particularly in domains with large datasets and complex data structures.
#science #physics
๐ก Donโt forget to like the video and subscribe for more edutainment content!
source
