Periodic Reporting for period 2 - EQUIPTNT (Enhanced quantum information processing targeting the near term)
Reporting period: 2023-04-01 to 2024-09-30
Progress in the fabrication, control and readout of systems consisting of a small number of qubits has catapulted quantum computation from a primarily theoretical pursuit into the lab. With the growing availability of small and relatively noise-free devices and the imminent arrival of larger next-generation equipment, new theoretical questions are being asked about the potential of quantum computation. In particular, one may ask if quantum computing devices may yield computational benefits even in the presence of noise or constraints on their size or geometrical layout of qubits.
The project EQUIPTNT seeks to characterize the computational capabilities of near-term quantum devices by studying their potential to yield disruptive boosts in information-processing power. It investigates and designs new quantum algorithms adapted to limited hardware: the aim here is to provide computational advances while maximizing noise-tolerance without placing excessive demands on experimental capabilities. It establishes trade-off relations between noise levels, computational power, and the amount and nature of available computational resources. EQUIPTNT also develops tools for simulating the quantum many-body dynamics of information-processing setups by classical algorithms. By doing so, the project pinpoints the origin of quantum advantage, and provides means for certifying the functionality of quantum hardware.
EQUIPTNT aims to establish new theoretical and algorithmic methods to address the question of ''best use'' for a given finite set of resources. Its interdisciplinary approach is expected to yield novel principles for the design, simulation and validation of quantum information processing protocols. Corresponding results have direct application to near-term quantum devices, providing insights into the architecture and use of schemes tailored towards specific experimental platforms.
The project EQUIPTNT seeks to characterize the computational capabilities of near-term quantum devices by studying their potential to yield disruptive boosts in information-processing power. It investigates and designs new quantum algorithms adapted to limited hardware: the aim here is to provide computational advances while maximizing noise-tolerance without placing excessive demands on experimental capabilities. It establishes trade-off relations between noise levels, computational power, and the amount and nature of available computational resources. EQUIPTNT also develops tools for simulating the quantum many-body dynamics of information-processing setups by classical algorithms. By doing so, the project pinpoints the origin of quantum advantage, and provides means for certifying the functionality of quantum hardware.
EQUIPTNT aims to establish new theoretical and algorithmic methods to address the question of ''best use'' for a given finite set of resources. Its interdisciplinary approach is expected to yield novel principles for the design, simulation and validation of quantum information processing protocols. Corresponding results have direct application to near-term quantum devices, providing insights into the architecture and use of schemes tailored towards specific experimental platforms.
Results obtained in the first reporting period advance our understanding of the potential of near- and intermediate-term quantum devices. Contributions encompass theoretical characterizations of their computational power, their resilience to noise and design imperfections, and an improved quantification of the difficulty of simulating, i.e. emulating their behavior by classical means. Advances with more direct practical relevance include novel protocols for fault-tolerantly generating long-range entanglement, the introduction of efficient decoders for quantum low-density parity check codes, new resource-efficient fault-tolerance constructions incorporating locality considerations, and novel methods for device characterization and validation derived from recent developments in quantum learning.
In the area of quantum complexity theory, a main finding is a new unconditional separation between similarly defined classical- and quantum circuit classes: It was shown that 3D-local, noisy constant-depth are computationally superior to constant-depth classical circuits with unbounded fan-in AND, OR and NOT gates. The corresponding quantum advantage proposal sheds new light on quantum gate teleportation, showing that this basic primitive has significant complexity-theoretic relevance.
New algorithms developed so far include hybrid (classical-quantum) algorithms for combinatorial optimization with improved guarantees on approximation ratios. These are obtained by combining a greedy algorithm with quantum approximate optimization. Furthermore, classical algorithms were developed to enable simulation of non-Gaussian dynamics. This extends the toolbox for studying many-body dynamics, providing new numerical methods to evaluate quantum computing proposals.
In the area of quantum complexity theory, a main finding is a new unconditional separation between similarly defined classical- and quantum circuit classes: It was shown that 3D-local, noisy constant-depth are computationally superior to constant-depth classical circuits with unbounded fan-in AND, OR and NOT gates. The corresponding quantum advantage proposal sheds new light on quantum gate teleportation, showing that this basic primitive has significant complexity-theoretic relevance.
New algorithms developed so far include hybrid (classical-quantum) algorithms for combinatorial optimization with improved guarantees on approximation ratios. These are obtained by combining a greedy algorithm with quantum approximate optimization. Furthermore, classical algorithms were developed to enable simulation of non-Gaussian dynamics. This extends the toolbox for studying many-body dynamics, providing new numerical methods to evaluate quantum computing proposals.
Together with the results obtained so far, the findings obtained until the end of EQUIPTNT are expected to provide a more fine-grained and broader understanding of the information-processing power of near-term quantum devices. In terms of computational complexity theory, this amounts to establishing new, previously unknown separations between classical and quantum computational complexity classes, as well as a more detailed characterization of when, e.g. noise leads to classical simulability.
Operationally useful advances are expected to include a variety of new methods and algorithms for simulating, certifying and error-mitigating many-body dynamics associated with near-term devices. Furthermore, it is expected that such results lead to simplified, e.g. more resource-efficient proposals for realizing existing, and possibly newly proposed quantum algorithms on imperfect hardware. By combining a variety of latest developments including in quantum learning theory and quantum error correction, these proposals are expected to go beyond the state of the art.
Operationally useful advances are expected to include a variety of new methods and algorithms for simulating, certifying and error-mitigating many-body dynamics associated with near-term devices. Furthermore, it is expected that such results lead to simplified, e.g. more resource-efficient proposals for realizing existing, and possibly newly proposed quantum algorithms on imperfect hardware. By combining a variety of latest developments including in quantum learning theory and quantum error correction, these proposals are expected to go beyond the state of the art.