Conference

Given the large number of proposed quantum machine learning (QML) algorithms, it is somewhat surprising that ideas from this field have not yet been extended to causal learning. While deep learning and generative machine learning models have taken centre stage in the industrial application of automated learning on classical data, it is nonetheless well known that these techniques don't reliably capture causal concepts, leading to significant performance vulnerabilities. Increasingly, classical ML experts are taking ideas from causal inference, a field traditionally limited to small data sets of low dimensionality, and injecting modern ML elements to create new algorithms that benefit from the best of both worlds. These hybrid classical approaches provide new opportunity to search for potential quantum advantage. In this talk I explore this new research direction and propose several new quantum algorithms for classical causal inference.

Quantum Chromodynamics (QCD) is the theory that describes the strong interaction. This interaction, mediated by gluons, is responsible for binding the quarks together to form protons, neutrons and other hadrons, as well as for holding the atomic nuclei together. At high energies, as quarks and gluons interact weakly, perturbation theory is a good tool for analysis. In contrast, the low-energy regime, which is crucial for understanding phenomena such as confinement, cannot be effectively treated with standard perturbative methods, calling for alternative approaches to be understood. In this talk, I will introduce the Curci-Ferrari model in Landau gauge as a framework to access the low-energy regime of QCD. The model modifies the action typically used for high-energy QCD by adding a mass term for the gluon field. This term models the behavior of the gluon propagator in numerical simulations, which behaves similarly to a massive field. The main strength of this model is that it quantitatively reproduces many results of numerical simulations for low-energy QCD, both at zero and at finite temperature and/or density, withminimal modeling. Additionally, the simplicity of the model enables well-controlled approximations that typically yield semi-analytic expressions. In the talk, I will present and discuss some results of the model, focusing on two-, three- and four-point correlation functions as examples of its performance.