Welcome to the Quantum Realm.

Enjoy today’s breakdown of news, research, & events within quantum.

🧮 Where VQE and QPE have fallen short, IBM Quantum has worked towards establishing the Krylov Quantum Diagonalization algorithm for estimating ground-state energies efficiently on up to 56 qubits. Plus, IonQ accelerates 2025 roadmap despite being in Q3 of 2024, cybersecurity concerns for quantum platforms themselves, and Universal Quantum’s ASIC chip for QPU integration.

🗓️UPCOMING

Sunday, July 28 | QTM-X Quantum Education Series 7 of 10

📰QUANTUM QUICK BYTES

🚀 IonQ unveils updated 2025 technical milestones with an emphasis on enterprise-grade quantum solutions: During a recent webinar, IonQ announced accelerated 2025 technical milestones focusing on performance, scalability, and enterprise-grade solutions during a recent webinar. The company committed to achieving above 99.9% native two-qubit gate performance and 99.999% logical two-qubit gate fidelity by the end of 2025 through architectural innovations and the introduction of barium qubits. IonQ will also be expanding its infrastructure through quantum data centers in the US and Switzerland to support industry demand.

🔒 Researchers emphasize the need to address cybersecurity risks to quantum computing platforms and not just their potential to break encryption: While the potential of quantum computers to break RSA encryption has long been a concern, researchers Adrian Colesa and Sorin Bolos highlight the overlooked risk that quantum computers are also vulnerable targets for cyberattacks. At the upcoming Black Hat USA 2024 conference this August, they will discuss what they have found on quantum vulnerabilities and the importance of securing quantum computing infrastructure. Their research identified attack vectors in quantum systems which illustrates the necessity of error correction and secure practices for SDKs and quantum circuits. With growing interest in quantum computing from industries like pharmaceuticals and finance, ensuring the security of these systems is of the utmost priority.

🔗 Universal Quantum develops new ASIC chip for QPU integration: Universal Quantum has developed the first commercial ASIC chip for integration into their quantum processing units which can enable efficient qubit control and connectivity at the million-qubit scale. The chip features UQConnect link technology and UQLogic microwave system, addresses the 'wiring problem', and improves quantum error correction through improved qubit connectivity. Dr. Mike Newman and CEO Sebastian Weidt spoke on the importance of scalable engineering solutions to meet the demand for million-qubit quantum computers.

🩺 Quantum computing has the potential to profoundly affect healthcare: Healthcare stands to benefit strongly from progress in quantum computing through solutions related to rapid drug discovery and precise diagnostics. The market for quantum computing in healthcare is projected to grow at a 42% CAGR from 2024 to 2032 through investments from both private and public sectors and the rise of over 500 startups. However, challenges such as a talent gap, high costs, and the need for noise and error correction must be addressed.

💻 IBM weighs in on the the current state of vulnerability for encryption and global finance: In a recent interview, Dr. Alessandro Curioni, Director of IBM Research at Zurich, discusses the both potential and risks of quantum computing. While quantum computers can perform calculations at unprecedented speeds, they can also encryption systems that protect critical infrastructure, financial markets, and personal data. Despite the timeline being uncertain, Curioni believes this breakthrough could happen by the end of the decade, with the United States and China leading the global race in quantum computing development due to their substantial investments.

☕️FRESHLY BREWED RESEARCH

DIAGONALIZATION OF LARGE MANY-BODY HAMILTONIANS ON A QUANTUM PROCESSOR

QUICK BYTE: Estimating low or ground-state energies in many-body systems is relevant across practical quantum computing applications but can be computationally challenging. The proposed Krylov Quantum Diagonalization algorithm combines a quantum subroutine with classical post-processing to effectively and scalably estimate these energies on current quantum devices.

PRE-REQS:

• A Krylov space is a subspace of a vector space that is generated by repeatedly applying a linear operator to a vector.

• The diagonalization of a matrix transforms a matrix such that all non-zero elements are on the main diagonal, and all off-diagonal elements are zero, ultimately simplifying linear operations.

SIGNIFICANCE: Estimating low or ground-state energies in many-body systems is relevant across various fields due to its computational complexity and potential solutions through quantum computing. In condensed matter physics, understanding the ground state energy is vital for predicting other physical properties of materials. In quantum chemistry, it is essential for predicting the outcome of chemical reactions, which is significant for understanding drug interactions with biological molecules in medicine.

Despite its importance, current methodologies for estimating ground-state energies have limitations. Variational quantum eigensolvers, though widely used, struggle with guaranteed convergence and require numerous iterations for parameter optimization which limits scalability. Quantum phase estimation, though more precise, requires substantial error correction due to its circuit depth.

This paper introduces a new methodology that combines a quantum subroutine with a classical post-processing step to make it scalable for general use. The proposed Krylov Quantum Diagonalization algorithm uses quantum computing to approximate the Hamiltonian's projection into a Krylov space. This results in a low-dimensional matrix that can be classically diagonalized to obtain approximate low-lying energy eigenstates. This approach is experimentally validated by estimating the ground-state energy of the Heisenberg model on a heavy-hexagonal lattice and achieving convergence on up to 56 qubits.

The advantage of the proposed model is its implementation on current devices without extensive circuit depth. Although noise is a factor, the model remains effective as long as the additional error term does not overwhelm the signal.

RESULTS:

• Validated by estimating the ground-state energy of the Heisenberg model on a heavy-hexagonal lattice and achieved convergence on up to 56 qubits, demonstrating scalability

• The KQD algorithm can be implemented on existing quantum devices without requiring extensive circuit depth.

• The model remains effective despite noise, as long as the noise does not significantly overwhelm the signal.

HONORABLE RESEARCH MENTIONS:

A new method for constructing a compact quantum random access memory circuit uses classical preprocessing to minimize quantum resources. The circuit orthogonalizes classical data, loads it into a quantum state using a parametrized quantum circuit, and then reverses the process to retrieve the original data. The efficiency was demonstrated through numerical simulations on handwritten digits and Iris datasets. —> link to Compact and classically preprocessed data-loading quantum circuit as a quantum random access memory

The use of sliding-window decoding for quantum low-density parity-check codes is explored to improve the memory lifetime of quantum systems. Unlike single-shot decoding, which can degrade code performance, sliding-window decoding corrects errors from previous rounds while leaving recent errors for future corrections which improves the logical memory lifetime without increasing complexity. —> link to Increasing memory lifetime of quantum low-density parity check codes with sliding-window noisy syndrome decoding

A quantum subroutine for computing distances between patterns is integrated into two quantum k-nearest neighbor algorithms and requires fewer qubits while maintaining the overall complexity. Benchmarking across thirteen datasets shows this approach reduces qubit requirements by at least 50% and, in some cases, improves performance. —> link to Benchmarking quantum versions of the kNN algorithm with a metric based on amplitude-encoded features

QuOp is introduced as a quantum operator representation for graph nodes using special unitary operators that eliminates the need for parameter training. QuOp uses the local topology of nodes to derive Hamiltonians from adjacency matrices and maps them into higher-dimensional Hilbert spaces for effective similarity scoring. Benchmarking against classical methods like GloVe and FastRP, QuOp demonstrates superior performance in maintaining effective similarity measures in graph structures, with promising applications in NLP and network anomaly detection. —> link to QuOp: A Quantum Operator Representation for Nodes

UNTIL TOMORROW.