Quantum Foundations

Out of the many lessons quantum mechanics seems to teach us, one is that it seems there are things we cannot experimentally have access to. There is, indeed, a fundamental limit to our ability to experimentally explore the world. In this work we accept this lesson as a fact and we build a general theory based on this principle. We start by assuming the existence of statements whose truth value is not experimentally accessible. That is, there is no way, not even in theory, to directly test if these statements are true or false. We further develop a theory in which experimentally accessible statements are a union of a fixed minimum number of inaccessible statements. For example, the value of truth of the statements a and b is not accessible, but the value of truth of the statement “a or b" is accessible. We do not directly assume probability theory, we exclusively define experimentally accessible and inaccessible statements and build on these notions using the rules of classical logic. We find that an interesting structure emerges. Developing this theory, we relax the logical structure, naturally obtaining a derivation of a constrained quasi-probabilistic theory rich in structure that we name theory of inaccessible information. Surprisingly, the simplest model of theory of inaccessible information is the qubit in quantum mechanics. Along the path for the construction of this theory, we characterise and study a family of multiplicative information measures that we call inaccessibility measures. arXiv:https://arxiv.org/abs/2305.05734

---

Zoom link https://pitp.zoom.us/j/91350754706?pwd=V1dVdGM3Zk9MNkp4VlpCYUoxbXg3UT09

We report a deterministic and exact protocol to reverse any unknown qubit-unitary and qubit-encoding isometry operations. To avoid known no-go results on universal deterministic exact unitary inversion, we consider the most general class of protocols transforming unknown unitary operations within the quantum circuit model, where the input unitary operation is called multiple times in sequence and fixed quantum circuits are inserted between the calls. In the proposed protocol, the input qubit-unitary operation is called 4 times to achieve the inverse operation, and the output state in an auxiliary system can be reused as a catalyst state in another run of the unitary inversion. This protocol only applies only for qubit-unitary operations, but we extend this protocol to any qubit-encoding isometry operations. We also present the simplification of the semidefinite programming for searching the optimal deterministic unitary inversion protocol for an arbitrary dimension presented by M. T. Quintino and D. Ebler [Quantum 6, 679 (2022)]. We show a method to reduce the large search space representing all possible protocols, which provides a useful tool for analyzing higher-order quantum transformations for unitary operations.

---

Zoom link https://pitp.zoom.us/j/92900413520?pwd=a1JqU1IzMVdSRGQreWJIbEFCT2hWUT09

Supermaps are higher-order transformations taking maps as input.  We consider quantum algorithms implementing supermaps for the input given by unknown Hamiltonian dynamics, which can be regarded as infinitely divisible unitary operations.  We first show a quantum algorithm that approximately but universally transforms black-box Hamiltonian dynamics into controlled Hamiltonian dynamics utilizing a higher-order transformation called neutralization.  Then, we present another universal algorithm that efficiently simulates linear transformations of any Hamiltonian consisting of a polynomial number of terms in system size, using only controlled-Pauli gates and time-correlated randomness.  This algorithm for implementing Hamiltonian supermaps is an instance of quantum functional programming, where the desired function is specified as a concatenation of higher-order quantum transformations. As examples, we demonstrate the simulation of negative time-evolution and time-reversal, and perform a Hamiltonian learning task.

References:

Q. Dong, S. Nakayama, A. Soeda and M. Murao, arXiv:1911.01645v3
T. Odake, Hlér Kristjánsson, A. Soeda M. Murao, arXiv:2303.09788

---

Zoom Link: https://pitp.zoom.us/j/94278362588?pwd=MGszYk9uN1A3K1RTOVhYSGpkL1FQdz09

In this informal talk, I shall give a short introduction to the field of higher-order quantum transformations, including its subfield of indefinite causal order. I shall discuss some of the motivations and important results in the field, current research directions and open problems, as well potential applications in quantum information processing.

I will begin by reviewing the general mathematical concept of representation. I will then show that representation theory is more generally applicable than one might expect, for example in quantum foundations, in quantum gravity and in machine learning. In those contexts, I will first talk about the notion of representation underlying phenomena of emergence in quantum gravity. I will then discuss how quantum reference frames might be viewed as representations. Finally, I will talk about how transformer models, such as GPT-4, construct representations of what they learn and, in that light, what it may take for machines to reach AGI or even consciousness.  

---

Zoom link: https://pitp.zoom.us/j/99349397588?pwd=T056VjZXRWZWT28zY3VOZHFDcmQ3QT09

One hundred years after Heisenberg’s Uncertainty Principle, the question of how to make simultaneous measurements of noncommuting observables lingers. I will survey one hundred years of measurement theory, which brings us to the point where we can formulate how to measure any set observables weakly and simultaneously and then concatenate such measurements continuously to determine what is a strong measurement of the same observables. The description of the measurements is independent of quantum states---this we call instrument autonomy---and even independent of Hilbert space---this we call the universal Instrument Manifold Program. But what space, if not Hilbert space? It’s a whole new world: the Kraus operators of an instrument live in a Lie-group manifold generated by the measured observables themselves. I will describe measuring position and momentum and measuring the three components of angular momentum, special cases where the instrument approaches asymptotically a phase-space boundary of the instrumental Lie-group manifold populated by coherent states; these special universal instruments structure any Hilbert space in which they are represented. In contrast, for almost all sets of observables other than these special cases, the universal instrument descends into chaos ... literally. This work was done with Christopher S. Jackson, whose genius and vision inform every aspect.

---

Zoom link https://pitp.zoom.us/j/94135518267?pwd=T2JOL21VaEcrY05KeG1SYTVYdHhxdz09

Tsirelson’s problem involves two agents, Alice and Bob, who apply measurements on the same quantum system, K. It asks whether commutation, i.e., independence of whether Alice or Bob measures first, is sufficient to conclude that Alice and Bob’s measurements can be factorised so that they act non-trivially only on distinct subsystems of K. In this talk, I will present a “fully quantum generalisation” of this problem, where Alice and Bob’s measurements are replaced by operations on K that may depend on additional quantum inputs and produce quantum outputs. As for Tsirelson’s original problem, it turns out that commutation indeed implies factorisation, provided that all relevant systems are finite-dimensional.

This is joint work with Ramona Wolf; preprint available at arXiv:2308.05792.

---

 Zoom link https://pitp.zoom.us/j/99031410183?pwd=MzVoQXpPSll6bFp1b1g3U2J4U21rZz09

Activation of Bell nonlocality refers to the phenomenon where some entangled mixed states that admit a local hidden variable model in the standard Bell scenario nevertheless reveal their nonlocal nature in more exotic measurement scenarios. It has recently been shown that by broadcasting the subsystems of a bipartite quantum state, one can activate Bell nonlocality and significantly improve noise tolerance bounds for device-independent and semi-device-independent entanglement certification, with a single copy of the state and local measurements. In this talk I will review the state of the art on activation of nonlocality and existence of local hidden variable models, introduce the broadcasting technique and outline several interesting results and research directions.

---

Zoom link  https://pitp.zoom.us/j/91214624839?pwd=dXNUOERLMldScjczemlaSUhFbFNWQT09

A number of no-go theorems suggest that theories upholding both unitarity and relativity must deny that events are absolute. I'll show how quantum causal influences allow us to articulate an attractive conception of relational events. This will lead us towards a precise, observer-independent reformulation of quantum theory, in which relational events emerge from causation.

Zoom link: https://pitp.zoom.us/j/98649944363?pwd=Y25NOUROelY2ZmFpVWhTZFErV2MwUT09

In 2020 our Oxford-based Quantinuum team performed Quantum Natural Language Processing (QNLP) on IBM quantum hardware [1, 2].  Key to having been able to achieve what is conceived as a heavily data-driven task, is the observation that quantum theory and natural language are governed by much of the same compositional structure -- a.k.a. tensor structure.  

Hence our language model is in a sense quantum-native, and we provide an analogy with simulation of quantum systems in terms of algorithmic speed-up [forthcoming].  Meanwhile we have made all our software available open-source, and with support [github.com/CQCL/lambeq].    

The compositional match between natural language and quantum extends to other domains than language, and argue that a new generation of AI can emerge when fully pushing this analogy, while exploiting the completeness of categorical quantum mechanics / ZX-calculus [3, 4, 5] for novel reasoning purposes that go hand-in-hand with modern machine learning.

[1]     B. Coecke, G. De Felice, K. Meichanetzidis and A. Toumi (2020) Foundations for Near-Term Quantum Natural Language Processing.  https://arxiv.org/abs/2012.03755

[2]     R. Lorenz, A. Pearson, K. Meichanetzidis, D. Kartsaklis and B. Coecke (2020) QNLP in Practice: Running Compositional Models of Meaning on a Quantum Computer.  https://arxiv.org/abs/2102.12846

[3]     B. Coecke and A. Kissinger (2017) Picturing Quantum Processes. A first course on quantum theory and diagrammatic reasoning.  Cambridge University Press.

[4]     B. Coecke, D. Horsman, A. Kissinger and Q. Wang (2021) Kindergarten quantum mechanics graduates (...or how I learned to stop gluing LEGO together and love the ZX-calculus).  https://arxiv.org/abs/2102.10984

[5]     B. Coecke and S. Gogioso (2022) Quantum in Pictures.  Quantinuum, 2023.

Zoom Link: https://pitp.zoom.us/j/92333285960?pwd=MlpJSklmMlVlUlRTTWhsNjc2T2Y4QT09

It is often the case that interaction with a quantum system does not simply occur between an initial point in time and a final one, but rather over many time steps. In such cases, an interaction at a given time step can have an influence on the dynamics of the system at a much later time. Just as quantum channels model dynamics between two time steps, quantum combs model the more general multi-time dynamics described above, and have accordingly found application in such fields as open quantum systems and quantum cryptography. In this talk, we will consider ensembles of combs indexed by a random variable, dubbed classical-quantum combs, and discuss how much can be learnt about said variable through interacting with the system. We characterise the amount of information gain using the comb min-entropy, an extension of the analogous entropic quantity for quantum states. With combs and the min-entropy in our toolbox, we turn to a number of applications largely inspired by Measurement-Based Quantum Computing (MBQC), including the security analysis of a specific Blind Quantum Computing protocol and some comments regarding learning causal structure.

Zoom Link: https://pitp.zoom.us/j/98315660866?pwd=cWU3RzB6SG9DOGIza1BqV1lqNklvQT09