Daniel Mills | LIP6 - Équipe QI

On-Chip Verified Quantum Computation with an Ion-Trap Quantum Processing Unit

Sun, 24 Nov 2024 00:00:00 +0000

Daniel Mills - Distributing circuits over heterogeneous, modular quantum computing network architectures

Wed, 27 Sep 2023 00:00:00 +0000

Distributing circuits over heterogeneous, modular quantum computing network architectures

Ce séminaire, donné par Daniel Mills, aura lieu le 27 September 2023, à 12:0. Il aura lieu en salle 105 Corridor 25-26.

Vous trouverez un plan du campus ici.

Résumé

This talk considers the compilation of quantum circuits to heterogeneous networks of quantum computing modules, sparsely connected via Bell states. A circuit too large to be implemented on any one module alone requires the insertion of operations, typically gate teleportation or qubit teleportation, consuming Bell states. These operations constitute a computational bottleneck and are likely to add more noise to the computation than operations performed within a module. We introduce several network architecture aware compilation techniques for transforming, through gate teleportation, a given quantum circuit to one implementable on a network of the aforementioned type, minimising the number of Bell states required. To do so we firstly extend the hypergraph approach of [Andres-Martinez & Heunen. 2019] to arbitrary network topologies. We introduce the use of Steiner trees to find efficient realisations of the entanglement sharing between modules, reusing already established connections as often as possible. Secondly, we extend the embedding techniques of [Wu, \textit{et al.} 2022], which allow for further entanglement reuse, to networks with more than two modules. We discuss how the seemingly incompatible approaches of embedding and of entanglement distribution with Steiner trees can be made to cooperate. Our proposals are implemented and benchmarked; the results confirming that, when orchestrated, the two approaches complement each other’s weaknesses.

The Born supremacy: quantum advantage and training of an Ising Born machine

Wed, 08 Jul 2020 00:00:00 +0000

The search for an application of near-term quantum devices is widespread. Quantum machine learning is touted as a potential utilisation of such devices, particularly those out of reach of the simulation capabilities of classical computers. In this work, we study such an application in generative modelling, focussing on a class of quantum circuits known as Born machines. Specifically, we define a subset of this class based on Ising Hamiltonians and show that the circuits encountered during gradient-based training cannot be efficiently sampled from classically up to multiplicative error in the worst case. Our gradient-based training methods use cost functions known as the Sinkhorn divergence and the Stein discrepancy, which have not previously been used in the gradientbased training of quantum circuits, and we also introduce quantum kernels to generative modelling. We show that these methods outperform the previous standard method, which used maximum mean discrepancy (MMD) as a cost function, and achieve this with minimal overhead. Finally, we discuss the ability of the model to learn hard distributions and provide formal definitions for ‘quantum learning supremacy’. We also exemplify the work of this paper by using generative modelling to perform quantum circuit compilation.

Methods for Classically Simulating Noisy Networked Quantum Architectures

Tue, 05 Nov 2019 00:00:00 +0000

As research on building scalable quantum computers advances, it is important to be able to certify their correctness. Due to the exponential hardness of classically simulating quantum computation, straight-forward verification via this means fails. However, we can classically simulate small scale quantum computations and hence we are able to test that devices behave as expected in this domain. This constitutes the first step towards obtaining confidence in the anticipated quantum-advantage when we extend to scales that can no longer be simulated. Real devices have restrictions due to their architecture and limitations due to physical imperfections and noise. In this paper we extend the usual ideal simulations by considering those effects. We provide a general methodology and framework for constructing simulations which emulate the physical system. These simulations should provide a benchmark for realistic devices and guide experimental research in the quest for quantum-advantage. To illustrate our methodology we give examples that involve networked architectures and the noise-model of the device developed by the Networked Quantum Information Technologies Hub (NQIT). For our simulations we use, with suitable modification, the classical simulator of Bravyi and Gosset while the specific problems considered belong to the Instantaneous Quantum Polynomial-time class. This class is believed to be hard for classical computational devices, and is regarded as a promising candidate for the first demonstration of quantum-advantage. We first consider a subclass of IQP, defined by Bermejo-Vega et al, involving two-dimensional dynamical quantum simulators, and then general instances of IQP, restricted to the architecture of NQIT.

The Born Supremacy: Quantum Advantage and Training of an Ising Born Machine

Tue, 25 Jun 2019 00:00:00 +0000

The search for an application of near-term quantum devices is widespread. Quantum Machine Learning is touted as a potential utilisation of such devices, particularly those which are out of the reach of the simulation capabilities of classical computers. In this work, we propose a generative Quantum Machine Learning Model, called the Ising Born Machine (IBM), which we show cannot, in the worst case, and up to suitable notions of error, be simulated efficiently by a classical device. We also show this holds for all the circuit families encountered during training. In particular, we explore quantum circuit learning using non-universal circuits derived from Ising Model Hamiltonians, which are implementable on near term quantum devices. We propose two novel training methods for the IBM by utilising the Stein Discrepancy and the Sinkhorn Divergence cost functions. We show numerically, both using a simulator within Rigetti’s Forest platform and on the Aspen-1 16Q chip, that the cost functions we suggest outperform the more commonly used Maximum Mean Discrepancy (MMD) for differentiable training. We also propose an improvement to the MMD by proposing a novel utilisation of quantum kernels which we demonstrate provides improvements over its classical counterpart. We discuss the potential of these methods to learn hard' quantum distributions, a feat which would demonstrate the advantage of quantum over classical computers, and provide the first formal definitions for what we call Quantum Learning Supremacy’. Finally, we propose a novel view on the area of quantum circuit compilation by using the IBM to `mimic’ target quantum circuits using classical output data only.

Information Theoretically Secure Hypothesis Test for Temporally Unstructured Quantum Computation (Extended Abstract)

Tue, 27 Feb 2018 00:00:00 +0000