- The Daily Qubit
- Posts
- The Daily Qubit

# The Daily Qubit

## 🗺️ It takes a village — industry professionals, academia, and innovators worldwide are making quantum happen. NSF grants provide the fuel, creative devices like the Quokka are delivering quantum education to the palm of your hand, and the patents abound.

# Welcome to the Quantum Realm.

🗺️ *It takes a village — industry professionals, academia, and innovators worldwide are making quantum happen. NSF grants provide the fuel, creative devices like the Quokka are delivering quantum education to the palm of your hand, and the patents abound.*

## 🗓️ UPCOMING

**Sunday, June 30 |** **QTM-X Quantum Education Series 5 of 10**** Quantum Hardware**

**Monday, July 1 |** TQN Quantum Safe Transition Working Group

## 📰 NEWS QUICK BYTES

🏆 Pasqal acheives a new order of magnitude, with no plans to slow down: Pasqal has successfully trapped over 1K atoms as a monumental step towards scalable neutral atom quantum processors. The precise manipulation of rubidium atoms at cryogenic temperatures allows for solving more complex optimization problems and performing intricate quantum simulations. Pasqal has set an ambitious goal to develop quantum processors with over 10K qubits by 2027.

🐨 Quokka brings quantum to everyone: Quokka is the world's first consumer product in quantum computing. This affordable, handheld quantum computer emulator created by Eigensystems allows users to learn and experiment with quantum algorithms, simulating a future 30-qubit fault-tolerant quantum computer. The goal of Quokka is to democratize access to quantum computing and to revolutionize STEM education with hands-on, interactive learning experiences.

🌐 ORCA pod to lead quantum networking: ORCA Computing, in collaboration with top quantum experts, has launched an R&D consortium to develop advanced multiplexing technologies for quantum networking. This will work towards scaling quantum computing for commercial applications by creating technologies essential for large-scale quantum data transfer.

⚛️ Quantum inception: Scientists at Forschungszentrum Jülich used a quantum annealer to successfully model the microscopic interactions of electrons in the quantum material 1T-TaS2. Not only does this demonstrate one of many practical applications within the field of materials science, but a deeper understanding of the quantum material itself could lead to the development of memory devices from this material. These devices, if implemented into a QPU, could reduce energy consumption for quantum computing systems, which would eliminate a major concern.

🧪 NSF funds quantum chemistry: Dr. Dmitri Babikov and his team are using quantum computers to study quantum chemistry, an area where classical computers fall short due to the exponential complexity of molecular interactions. Supported by an $800,000 NSF grant, Babikov's research at Marquette and Los Alamos National Laboratory involves hands-on use of quantum hardware like D-Wave's Quantum Annealer and IBM-Q machines to develop algorithms suited for quantum computing to advance understanding in quantum chemistry and molecular dynamics.

📚 NSF funds hybrid computing: Dr. Peng Guo of Dakota State University has received a prestigious LEAPS-MPS grant from the NSF to support his research in quantum physics. His work focuses on improving predictions of few-nucleon reactions using AI, machine learning, and quantum computing, as well as exploring PT-symmetric quantum materials.

🔬 EVG and Fraunhofer innovate quantum integration: EV Group and Fraunhofer IZM-ASSID have formed a partnership to work towards advanced bonding and debonding technologies for CMOS and heterogeneous integration, including quantum computing. Fraunhofer's new Center for Advanced CMOS and Heterointegration Saxony in Dresden has installed EVG's EVG850 DB UV laser debonding and cleaning system to complement their research in 300-mm 3D wafer-level integration and semiconductor processes. This collaboration can accelerate the development of quantum systems and advanced semiconductor technologies by providing state-of-the-art in-house capabilities and fostering innovation in 3D device integration.

## How many qubits was today's newsletter? |

## ☕️ FRESHLY BREWED RESEARCH

Reduction of finite sampling noise in quantum neural networks: A technique called variance regularization is used to mitigate finite sampling noise in quantum neural networks to improve training efficiency, since the cost of repeated circuit evaluations is high. Empirical benchmarks, including regression tasks and potential energy surface interpolation, are included to show that this method reduces noise by an order of magnitude and effectively makes QNNs more practical and efficient for use on current quantum hardware. Breakdown here.

Comprehensive characterization of three-qubit Grover search algorithm on IBM's 127-qubit superconducting quantum computers: A three-qubit Grover search algorithm is implemented on IBM’s 127-qubit superconducting quantum computer to analyze its performance across noise-free, noisy, and real quantum hardware environments. While the algorithm unsurprisingly shows potential for significant speedup in unstructured database searches, there are clear practical challenges such as noise and hardware limitations. Breakdown here.

Path-Breaking Directions in Quantum Computing Technology: A Patent Analysis with Multiple Techniques: The analysis of the evolutionary patterns of quantum computing technology is conducted using patent data and three stages are defined: emerging (1992-2008), growth (2009-2017), and maturity (2018-2022). Since 2016, there has been a surge in patents which is a key indicator that the technology is in its maturity stage. The research provides insights into the complex interconnections and evolutionary trajectories of quantum computing topics, which can inform strategic planning, innovation management, and policy-making.

A benchmarking study of quantum algorithms for combinatorial optimization: A benchmarking study of three quantum algorithms for combinatorial optimization is conducted. The algorithms include measurement-feedback coherent Ising machines (MFB-CIM), discrete adiabatic quantum computation (DAQC), and the Dürr-Høyer quantum minimum finding algorithm (DH-QMF) based on Grover's search. Their performance is compared using Max cut problems. The MFB-CIM is shown to demonstrate significant performance advantages with sub-exponential scaling and outperforms DAQC and DH-QMF.

Quantum computer specification for nuclear structure calculations: The variational quantum eigensolver algorithm is used to solve nuclear structure problems. The correlation energy of Helium-6 is calculated using a full-term unitary-paired-coupled-cluster-doubles ansatz on a quantum computer simulator. Minimal specifications of 5 ms coherence times and 10^-4 quantum errors are required to reproduce state-vector results within an 8% discrepancy, which shows the potential for VQE calculations on slightly noisy quantum computers without the need for quantum error correction.

Quantum Extreme Learning of molecular potential energy surfaces and force fields: Quantum extreme learning machines are used to predict molecular potential energy surfaces and force fields. QELM is shown to efficiently learn and predict these properties with high accuracy for molecules like lithium hydride, water, and formamide using a minimal quantum resource setup suitable for NISQ devices. QELM outperforms traditional and other quantum methods in terms of both accuracy and resource efficiency.

## UNTIL TOMORROW.

### BREAKDOWN

**Reduction of finite sampling noise in quantum neural networks**

🔍️ **SIGNIFICANCE:**

Evaluating expectation values in QNNs requires many repeated circuit executions which will inevitably introduce noise, even on error-free quantum computers. This noise hampers the training efficiency and accuracy of QNNs. The authors introduce variance regularization as a technique used to reduce the variance of expectation values during QNN training without additional circuit evaluations.

🧪 **METHODOLOGY:**

A parameterized quantum circuit is used to encode input data using rotation gates and manipulate quantum states through a series of one and two-qubit gates.

Variance regularization is introduced by adding a term to the loss function that minimizes the variance of the output at a set of points, making it so additional circuit evaluations are not required.

The QNN can then optimize both fitting and variance reduction simultaneously. Optimization techniques include gradient-based optimization methods to adjust the parameters of the PQC and cost operators.

The number of shots are adjusted based on the relative standard deviation of the fitting loss to effectively optimize the number of shots required during gradient evaluations.

To demonstrate whether or not the variance regularization is effective, empirical benchmarks on the regression of multiple functions and the potential surface energy of water are included.

📊 **OUTCOMES & OUTLOOK:**

Variance regularization is shown to significantly reduce the finite sampling noise in QNNs. More quantitatively, the technique reduces the variance by an order of magnitude which greatly decreases the noise level in QNN outputs.

By lowering the variance, fewer evaluations of gradient circuits are needed which effectively speeds up the training process.

The combined reduction in noise, reduction in gradient circuit evalulation, and quicker training makes this method both feasible and beneficial on current quantum devices.

^{Source:}^{ Kreplin, David A. and Roth, Marco. Reduction of finite sampling noise in quantum neural networks.}^{ Quantum}^{. (2024). }^{https://doi.org/10.22331/q-2024-06-25-1385}

### BREAKDOWN

**Comprehensive characterization of three-qubit Grover search algorithm on IBM's 127-qubit superconducting quantum computers**

🔍️ **SIGNIFICANCE:**

When searching for a marked item in an unstructured database, Grover's algorithm has shown a quadratic speedup over classical algorithms. This is relevant for various applications in cryptography, optimization, and database search. Previous implementations of Grover’s algorithm have been conducted on smaller systems or trapped ion systems. This study specifically uses IBM's 127-qubit superconducting quantum computer to analyze the capability and practicality of implementing Grover’s algorithm on today’s quantum devices.

🧪 **METHODOLOGY:**

Qubits were prepared in a superposition state using Hadamard gates and the target states were identified and marked by applying phase oracles to modify the amplitude of the marked states.

The probability amplitude of the marked states was increased using Grover operators and the final states were measured to obtain the search results.

The algorithm was implemented with all eight possible single-result oracles and nine two-result oracles. Five quantum state tomography experiments were conducted to evaluate the behavior and efficiency of the algorithm under different conditions, including noise-free, simulated noisy environments, and real quantum hardware.

📊 **OUTCOMES & OUTLOOK:**

The average algorithm success probability for single-solution scenarios was 78.39% in noisy environments and 51.19% on IBM's real quantum computers. For two-solution scenarios, the ASP was 84.44% in noisy environments and 64.44% on real quantum hardware.

The average squared statistical overlap for single-solution scenarios was 82.358% in noisy environments and 73.12% on real quantum computers. For two-solution scenarios, the SSO was 84.03% in noisy environments and 63.10% on real quantum hardware.

These results demonstrate that while the Grover search algorithm shows strong performance in idealized conditions, its effectiveness is significantly impacted by noise and hardware limitations in real-world quantum computing environments. NISQ computers have potential for practical implementations but progress in error mitigation and quantum hardware reliability are strongly recommended.

^{Source:}^{ M. AbuGhanem. Comprehensive characterization of three-qubit Grover search algorithm on IBM's 127-qubit superconducting quantum computers. }^{arXiv quant-ph.}^{ (2024). }^{https://arxiv.org/abs/2406.16018v1}

## Support Science

Waking up before the world to dive into the quantum realm isn't just our job—it's our calling. And we're dreaming big with exclusive content for our community. If our work lights up your day, consider showing some love. Your support unlocks worlds—seen and unseen.

Interested in collaboration or promoting your company, product, job, or event to the quantum computing community? Reach out to us at **[email protected]**