Florian Krellner

Florian Krellner

Auto-generated profile, based on publicly available information

Florian Krellner

SAP SE Walldorf Germany

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 June 2024 in ACM Transactions on Quantum Computing
Verified
Journal Article
Open access
CRCarlos A. Riofrio
Carlos A. Riofrio
OMOliver Mitevski
Oliver Mitevski
CJCaitlin Jones
Caitlin Jones
FKFlorian Krellner
Florian Krellner
AVAleksandar Vuckovic
Aleksandar Vuckovic
ALAndre Luckow
Andre Luckow
Quantum generative modeling is a growing area of interest for industry-relevant applications. This work systematically compares a broad range of techniques to guide quantum computing practitioners when deciding which models and methods to use in their applications. We compare fundamentally different architectural ansatzes of parametric quantum circuits: (1) A continuous architecture, which produces continuous-valued data samples, and (2) a discrete architecture, which samples on a discrete grid. We also compare the performance of different data transformations: the min-max and the probability integral transforms. We use two popular training methods: (1) quantum circuit Born machines (QCBM), and (2) quantum generative adversarial networks (QGAN). We study their performance and tradeoffs as the number of model parameters increases, with a baseline comparison of similarly trained classical neural networks. The study is performed on six low-dimensional synthetic and two real financial data sets. Our two key findings are that: (1) For all data sets, our quantum models require similar or fewer parameters than their classical counterparts. In the extreme case, the quantum models require two orders of magnitude less parameters. (2) We empirically find that a variant of the discrete architecture, which learns the copula of the probability distribution, outperforms all other methods.

7 February 2025 in arXiv
Verified
Preprint
Open access
PDF file
YVYelyzaveta Vodovozova
Yelyzaveta Vodovozova
AAAbhishek Awasthi
Abhishek Awasthi
CJCaitlin Jones
Caitlin Jones
JDJoseph Doetsch
Joseph Doetsch
KWKaren Wintersperger
Karen Wintersperger
CRCarlos A. Riofrío
Carlos A. Riofrío
Optimization is a crucial task in various industries such as logistics, aviation, manufacturing, chemical, pharmaceutical, and insurance, where finding the best solution to a problem can result in significant cost savings and increased efficiency. Tensor networks (TNs) have gained prominence in recent years in modeling classical systems with quantum-inspired approaches. More recently, TN generative-enhanced optimization (TN-GEO) has been proposed as a strategy which uses generative modeling to efficiently sample valid solutions with respect to certain constraints of optimization problems. Moreover, it has been shown that symmetric TNs (STNs) can encode certain constraints of optimization problems, thus aiding in their solution process. In this work, we investigate the applicability of TN- and STN-GEO to an industry relevant problem class, a multi-knapsack problem, in which each object must be assigned to an available knapsack. We detail a prescription for practitioners to use the TN-and STN-GEO methodology and study its scaling behavior and dependence on its hyper-parameters. We benchmark 60 different problem instances and find that TN-GEO and STN-GEO produce results of similar quality to simulated annealing.

29 July 2025 in arXiv
Verified
Preprint
Open access
PDF file
FKFlorian Krellner
Florian Krellner
AAAbhishek Awasthi
Abhishek Awasthi
NKNico Kraus
Nico Kraus
SBSarah Braun
Sarah Braun
MPMichael Poppel
Michael Poppel
DPDaniel Porawski
Daniel Porawski
This study evaluates the performance of a quantum-classical metaheuristic and a traditional classical mathematical programming solver, applied to two mathematical optimization models for an industry-relevant scheduling problem with autonomous guided vehicles (AGVs). The two models are: (1) a time-indexed mixed-integer linear program, and (2) a novel binary optimization problem with linear and quadratic constraints and a linear objective. Our experiments indicate that optimization methods are very susceptible to modeling techniques and different solvers require dedicated methods. We show in this work that quantum-classical metaheuristics can benefit from a new way of modeling mathematical optimization problems. Additionally, we present a detailed performance comparison of the two solution methods for each optimization model.

26 March 2024 in arXiv
Verified
Preprint
Open access
PDF file
CRCarlos A. Riofrío
Carlos A. Riofrío
OMOliver Mitevski
Oliver Mitevski
CJCaitlin Jones
Caitlin Jones
FKFlorian Krellner
Florian Krellner
AVAleksandar Vučković
Aleksandar Vučković
ALAndre Luckow
Andre Luckow
Quantum generative modeling is a growing area of interest for industry-relevant applications. With the field still in its infancy, there are many competing techniques. This work is an attempt to systematically compare a broad range of these techniques to guide quantum computing practitioners when deciding which models and techniques to use in their applications. We compare fundamentally different architectural ansatzes of parametric quantum circuits used for quantum generative modeling: 1. A continuous architecture, which produces continuous-valued data samples, and 2. a discrete architecture, which samples on a discrete grid. We compare the performance of different data transformations: normalization by the min-max transform or by the probability integral transform. We learn the underlying probability distribution of the data sets via two popular training methods: 1. quantum circuit Born machines (QCBM), and 2. quantum generative adversarial networks (QGAN). We study their performance and trade-offs as the number of model parameters increases, with the baseline of similarly trained classical neural networks. The study is performed on six low-dimensional synthetic and two real financial data sets. Our two key findings are that: 1. For all data sets, our quantum models require similar or fewer parameters than their classical counterparts. In the extreme case, the quantum models require two of orders of magnitude less parameters. 2. We empirically find that a variant of the discrete architecture, which learns the copula of the probability distribution, outperforms all other methods.

3 August 2024 in arXiv
Verified
Preprint
Open access
PDF file
AAAbhishek Awasthi
Abhishek Awasthi
NKNico Kraus
Nico Kraus
FKFlorian Krellner
Florian Krellner
DZDavid Zambrano
David Zambrano
This work is a benchmark study for quantum-classical computing method with a real-world optimization problem from industry. The problem involves scheduling and balancing jobs on different machines, with a non-linear objective function. We first present the motivation and the problem description, along with different modeling techniques for classical and quantum computing. The modeling for classical solvers has been done as a mixed-integer convex program, while for the quantum-classical solver we model the problem as a binary quadratic program, which is best suited to the D-Wave Leap's Hybrid Solver. This ensures that all the solvers we use are fetched with dedicated and most suitable model(s). Henceforth, we carry out benchmarking and comparisons between classical and quantum-classical methods, on problem sizes ranging till approximately 150000 variables. We utilize an industry grade classical solver and compare its results with D-Wave Leap's Hybrid Solver. The results we obtain from D-Wave are highly competitive and sometimes offer speedups, compared to the classical solver.