PD Dr.

Lorenz, Jeanette Miriam

Filters

6 1
Reset filters

Settings
Now showing 1 - 7 of 7
No Thumbnail Available
Publication

Optimizing hyperparameters using the geometric difference

2023 , Egginger, Sebastian , Sakhnenko, Alona , Runge, Xiomara , Lorenz, Jeanette Miriam

Quantum kernel methods (QKM) are a promising method in Quantum machine learning (QML) thanks to the guarantees connected to them. Their accessibility for analytic considerations also opens up the possibility of prescreening datasets based on their potential for a quantum advantage. To do so, earlier works developed the geometric difference, which can be understood as a closeness measure between two kernel-based ML approaches, most importantly between a quantum kernel and classical kernel. This metric links the quantum and classical model complexities. Therefore, it raises the question of whether the geometric difference, based on its relation to model complexity, can be a useful tool in evaluations other than the potential for quantum advantage.

View
No Thumbnail Available
Publication

Pooling Techniques in Hybrid Quantum-Classical Convolutional Neural Networks

2023 , Monnet, Maureen , Gebran, Hanady , Matic-Flierl, Andrea , Kiwit, Florian , Schachtner, Balthasar , Bentellis, Amine , Lorenz, Jeanette Miriam

Quantum machine learning has received significant interest in recent years, with theoretical studies showing that quantum variants of classical machine learning algorithms can provide good generalization from small training data sizes. However, there are notably no strong theoretical insights about what makes a quantum circuit design better than another, and comparative studies between quantum equivalents have not been done for every type of classical layers or techniques crucial for classical machine learning. Particularly, the pooling layer within convolutional neural networks is a fundamental operation left to explore. Pooling mechanisms significantly improve the performance of classical machine learning algorithms by playing a key role in reducing input dimensionality and extracting clean features from the input data. In this work, an in-depth study of pooling techniques in hybrid quantum-classical convolutional neural networks (QCCNNs) for classifying 2D medical images is performed. The performance of four different quantum and hybrid pooling techniques is studied: mid-circuit measurements, ancilla qubits with controlled gates, modular quantum pooling blocks and qubit selection with classical postprocessing. We find similar or better performance in comparison to an equivalent classical model and QCCNN without pooling and conclude that it is promising to study architectural choices in QCCNNs in more depth for future applications.

View
No Thumbnail Available
Publication

Quantencomputing für industrielle Softwareanwendungen

2021 , Lorenz, Jeanette Miriam

Obwohl Quantencomputer aktuell noch nicht leistungsstark genug sind um in der Industrie breite Anwendung zu finden, wird vorhergesagt, dass Quantencomputing in vielen Feldern innerhalb der nächsten 10 Jahre zu tiefgehenden Veränderungen führen wird. Dabei haben Quantencomputer aufgrund ihrer quantenmechanischen Eigenschaften das Potential sowohl klassische Software-Algorithmen deutlich zu beschleunigen, wie auch komplett neue Problemklassen zu erschließen. Beispiele für Anwendungen sind eine effizientere Simulation von Molekülen für Katalysatoren, Enzyme oder Medikamenten, Simulation von Batterien im Energiesektor oder für das autonome Fahren, oder auch die Lösung komplexer Optimierungsprobleme beispielsweise im Logistiksektor. Auch im Bereich der künstlichen Intelligenz kann das Quantencomputing zu tiefgreifenden Veränderungen führen, beispielsweise durch ein deutlich effizienteres Training von neuronalen Netzen. Bei diesen vielfältigen Anwendungsmöglichkeiten und den zugleich zu erwartenden disruptiven Veränderungen ist es notwendig frühzeitig zu gewährleisten, dass Quantencomputing verlässlich und für die Endanwender sicher eingesetzt werden kann. In diesem Vortrag wird zunächst erläutert, warum angenommen wird, dass Quantencomputing einige Softwarealgorithmen deutlich effizienter gestalten kann, und sodann einige der Anwendungsfelder und Anwendungsbeispiele diskutiert. Abschließend wird ein Ausblick über die zu erwartende Entwicklung des Quantencomputings in den nächsten Jahren gegeben.

View
No Thumbnail Available
Publication

Quantum-enhanced AI in medicine

2023 , Lorenz, Jeanette Miriam

Quantum computing is predicted as distruptive technologies with the capabilties to analyize complex patterns in data. The medical sector is a challenging field for applying artifical intelligence methods due to different reasons, but one of them being the limited amount of training data available. This talk describes how quantum computing might be able to address some of the open challenges in the sector of digital health, as in particular for the case where only limited training data is available.

View
No Thumbnail Available
Publication

Machine learning and simulation on NISQ devices

2022 , Sakhnenko, Alona , Lorenz, Jeanette Miriam

Our team is investigating many fields, which hold a potential for a practical quantum advantage. In this presentation, we show-case some of our recent achievements within the realm of quantum simulation and quantum machine learning. Specifically, we presented some of our results within Munich Quantum Valley project, such as a recent publication in Noise impact investigation for VQE algorithm, results in Quantum reinforcement learning and current state-of-the-art in Quantum kernel methods. A recent publication in Quantum Convolutional NN was presented as well.

View
No Thumbnail Available
Publication

Quantum Neural Networks under Depolarization Noise: Exploring White-Box Attacks and Defenses

2023 , Winderl, David , Franco, Nicola , Lorenz, Jeanette Miriam

Leveraging the unique properties of quantum mechanics, Quantum Machine Learning (QML) promises computational breakthroughs and enriched perspectives where traditional systems reach their boundaries. However, similarly to classical machine learning, QML is not immune to adversarial attacks. Quantum adversarial machine learning has become instrumental in highlighting the weak points of QML models when faced with adversarial crafted feature vectors. Diving deep into this domain, our exploration shines light on the interplay between depolarization noise and adversarial robustness. While previous results enhanced robustness from adversarial threats through depolarization noise, our findings paint a different picture. Interestingly, adding depolarization noise discontinued the effect of providing further robustness for a multi-class classification scenario. Consolidating our findings, we conducted experiments with a multi-class classifier adversarially trained on gate-based quantum simulators, further elucidating this unexpected behavior.

View
No Thumbnail Available
Publication

Quantum-classical convolutional neural networks in radiological image classification

2022 , Matic, Andrea , Monnet, Maureen , Schachtner, Balthasar , Lorenz, Jeanette Miriam , Messerer, Thomas

Quantum machine learning is receiving significant attention currently, but its usefulness in comparison to classical machine learning techniques for practical applications remains unclear. However, there are indications that certain quantum machine learning algorithms might result in improved training capabilities with respect to their classical counterparts - which might be particularly beneficial in situations with little training data available. Such situations naturally arise in medical classification tasks. Within this paper, different hybrid quantum-classical convolutional neural networks (QCCNN) with varying quantum circuit designs and encoding techniques are proposed. They are applied to two- and three-dimensional medical imaging data, e.g. featuring different, potentially malign, lesions in computed tomography scans. The performance of these QCCNNs is already similar to the one of their classical counterparts therefore encouraging further studies towards the direction of applying these algorithms within medical imaging tasks.

View