Explore chapters and articles related to this topic
Teamplay
Published in Volker Knecht, AI for Physics, 2023
As a result of these quantum effects, the computing power of quantum computers can increase exponentially with the number of qubits. This applies to various algorithms, including quantum neural networks; these artificial NN models are based on the principles of quantum mechanics.
Healthcare System 4.0 Driven by Quantum Computing and Its Use Cases
Published in Thiruselvan Subramanian, Archana Dhyani, Adarsh Kumar, Sukhpal Singh Gill, Artificial Intelligence, Machine Learning and Blockchain in Quantum Satellite, Drone and Network, 2023
The possible future scopes in a QC-based healthcare system are as below: Hospitals on Clouds: They are voluminous in nature and also need temperatures about 15 millidegrees above absolute zero to operate. They require huge computation units those require more space naturally. Because of these features, they are hard to install at the required location rather than making them available over the cloud.Cognizance of Quantum: It includes human brain like unit, language skills, decision-making ability, memorizing capability, and analytical abilities in task doing. Quantum probabilities play a major role in all these activities of cognizance of quantum. This aspect plays a vital role in healthcare in order to think like the human brain in tackling health problems and providing their solutions.Cryptography via Quantum: It aims the development of encryption methods that are stronger than the traditional ones available at present. Quantum cryptography gives scope for the development of such methods to strengthen security further. They draw that strength from the quantum mechanics properties of particles. If anyone wants to grab encoded data, the quantum state captures that attempt. It shows that attempt by changing its state immediately. This facilitates noticing such an attempt. Then it can be prevented from taking place. The healthcare system with this option will provide for high security of medical data.Quantum Neural Network (QNN): Neural networks can be extended to a new domain called Quantum-oriented Neural Networks. They have features of ordinary neural networks along with embedded quantum principles. This will provide an opportunity to develop more efficient algorithms. These algorithms help to extend networks, memory devices, and control systems automation for the quantum environment. This will enable for automation of the overall healthcare system.
The unified effect of data encoding, ansatz expressibility and entanglement on the trainability of HQNNs
Published in International Journal of Parallel, Emergent and Distributed Systems, 2023
Muhammad Kashif, Saif Al-Kuwari
Quantum neural networks (QNNs) are a key focus in Quantum Machine Learning (QML), inspired by the success of classical neural networks (NNs). However, the practical usability of QNNs is hindered by barren plateaus (BP) problem, where parameter gradients become extremely small as system size increases, rendering QNNs difficult to train. The primary components of QNNs (data encoding, ansatz expressibility, and entanglement) have already been individually studied, but understanding their combined effects is crucial for the practical applicability of QNNs. In this paper, we propose a framework to empirically investigate the holistic effect of all the aforementioned components of QNNs for a practical application namely: multi-class classification. In a practical setting, because of the limitations of noisy intermediate-scale quantum devices, hybrid quantum neural networks (HQNNs) are widely being used to explore the potential quantum advantage of QNNs. Since HQNNs completely replicate the general QNN architecture (with some classical input pre- and post-processing), the analysis of the quantum parts of HQNN can be directly applicable to QNNs. Typically, HQNNS consists of the following: (1) input dimensionality reduction, (2) qubit initialization, (3) data encoding (classical to quantum feature mapping), (4) quantum ansatz (parameterized quantum circuit), (5) qubit measurements and (6) dense classical neuron layer to post-process the qubit measurement results and get the output.