The Speakers

Affiliation: Global Technology Applied Research, JPMorganChase  

Title:  Quantum Approximate Optimization of Integer Graph Problems and Surpassing Semidefinite Programming for Max-k-Cut

Abstract: Quantum algorithms for binary optimization problems have been the subject of extensive study. However, the application of quantum algorithms to integer optimization problems remains comparatively unexplored. In this paper, we study the Quantum Approximate Optimization Algorithm (QAOA) applied to integer problems on graphs, with each integer variable encoded in a qudit. We derive a general iterative formula for depth-p QAOA expectation on high-girth d-regular graphs of arbitrary size. The cost of evaluating the formula is exponential in the QAOA depth p but does not depend on the graph size. Evaluating this formula for Max-k-Cut problem for p≤4, we identify parameter regimes (k=3 with degree d≤10 and k=4 with d≤40) in which QAOA outperforms the Frieze-Jerrum semi-definite programming (SDP) algorithm, which provides the best worst-case guarantee on the approximation ratio. To strengthen the classical baseline, we introduce a new heuristic algorithm based on the degree-of-saturation that achieves strong results on the \texttt{GSet} benchmark with quasi-linear runtime in the number of edges. It empirically outperforms both the Frieze-Jerrum algorithm and shallow-depth QAOA on regular graphs. Nevertheless, we provide numerical evidence that QAOA may overtake this heuristic at depth p≤20. Our results show that moving beyond binary to integer optimization problems can open up new avenues for quantum advantage.

https://arxiv.org/abs/2602.05956

Affiliation: NVIDIA

Title:  AI-Accelerated Quantum Optimization from Adaptive Circuits to Distributed Combinatorial Problem Solving

Abstract: Quantum optimization offers a promising path toward tackling complex combinatorial problems, yet designing effective quantum algorithms and scaling them to larger problems remain major challenges. In this talk, I will explore how generative AI can help bridge this gap through QAOA-GPT and DQAOA-GPT. These frameworks use CUDA-Q and AI to assist in designing adaptive quantum optimization circuits and extend them toward distributed quantum computing, where larger problems can be addressed across multiple quantum resources. Together, this work points toward a future in which AI and quantum computing work hand in hand to automate algorithm design, improve scalability, and accelerate the development of practical quantum optimization solutions.

Bio: Dr Marwa Farag is a Senior Quantum Algorithm Engineer at NVIDIA, working on hybrid quantum-classical algorithm and  AI‑assisted quantum algorithm design for scalable and real‑world applications.

Prior to joining NVIDIA, Dr. Farag led quantum computing research at Ford Motor Company, focusing on quantum and hybrid approaches for advanced materials simulation. She earned her Ph.D. in theoretical chemistry and computational modeling from the University of Murcia and subsequently held postdoctoral research positions at the Max Planck Institute, the University of Groningen, and the University of Southern California.

During her doctoral and postdoctoral research, she developed quantum chemical methods for electronic structure theory and the dynamics of chemical systems.

Affiliation: Quantum Artificial Intelligence Laboratory, NASA & USRA Research Institute for Advanced Computer Science

Title:  Quantum optimization heuristics for 100-1000 semi-protected qubits

Abstract: Early fault-tolerant, megaquop-scale machines are now arriving, and with them a fast-growing body of experimentation on computing use cases built around a few hundred imperfect logical qubits. Because co-design at this scale is played under the constraints of error handling, such platforms favor noise-robust or noise-exploiting heuristics for quantum optimization and quantum machine learning over more elaborate algorithms with provable advantage, making these heuristics a natural focus for early application development.

Drawing on recent concrete examples (e.g. [1-4] and others), we will examine the three gaps that are left over from the NISQ era and how algorithm design, hybridization, application performance, and hardware benchmarking are all shifting in response to the technological push to fault-tolerance and toward a new set of quantum-engineering capabilities.

[1] Quantum approximate optimization via noise-directed adaptive warm-starting (arXiv:2607.09368 - 2026).

[2] Setting angles in quantum approximate optimization at utility-scale. arXiv:2606.05311 (2026).

[3] Noise-directed adaptive remapping for integer optimization: from qubits to (encoded) qudits. arXiv:2506.05608 (2025).  

[4] Near-term application engineering challenges in emerging superconducting qudit processors. arXiv:2506.05608 (2025)

Bio: Davide Venturelli, PhD, is a USRA Distinguished Fellow and the Associate Director for Quantum at the Universities Space Research Association (USRA) Research Institute for Advanced Computer Science (RIACS). He concurrently serves as a Senior Research Scientist contractor at the NASA Quantum AI Laboratory (QuAIL) since its foundation in 2012. As a Principal Investigator for DOD, NSF, and DOE projects, including co-leading the Quantum Computing Use Case group at Fermilab under the National Quantum Initiative SQMS center, he directed benchmarking and co-design efforts on state-of-the-art quantum processors (superconducting and atomic modalities).

Affiliation: University of Tennessee Chattanooga

Title:  Hybrid Quantum-Classical Algorithms for Power Grid Optimization

Abstract: Alternating Current Optimal Power Flow Unit Commitment (AC-OPF-UC) is a difficult mixed-integer nonlinear optimization problem that combines binary generator commitment decisions with nonconvex continuous AC power-flow constraints. In this work, we investigate whether hybrid quantum-classical variational algorithms can improve the solution of single-period AC-OPF-UC relative to classical approaches. To the best of our knowledge, this is the first study to directly evaluate quantum or hybrid quantum-classical algorithms for the full AC-OPF-UC problem. We consider two candidate algorithms for improving AC-OPF-UC solution quality relative to purely classical methods on ideal quantum hardware. The first applies QAOA directly to a fully discretized formulation of the problem, with equality and inequality constraints incorporated through penalty terms and slack variables. Although conceptually straightforward, this approach requires a prohibitively large number of qubits even for small instances. The second, qubit-efficient approach encodes only the binary generator status variables on a quantum computer, while optimizing the continuous power-flow variables classically for each sampled bitstring. We benchmark this method on randomly generated AC-OPF-UC instances with 5 to 13 generators and compare it against SCIP, SMAC, and uniform random sampling. Our simulations show that the qubit-efficient hybrid method does not outperform uniform sampling. These results suggest that in order to establish potential advantage of the variational hybrid strategy considered here over the best classical algorithms, if any, much larger system sizes (25+ generators) need to be tested, which is beyond our computational capacity. Alternatively, different approaches, such as quantum versions of branch-and-bound methods, may be more promising.

https://arxiv.org/abs/2607.15543

Bio: Igor Gaidai is an Assistant Research Professor in the Quantum Center at the University of Tennessee in Chattanooga. He received his Ph.D. in Quantum Chemistry from Marquette University and also holds an M.S. in Bioinformatics and a B.S. in Computer Science. His current research focuses on theoretical quantum algorithms, particularly for combinatorial optimization and physics-related applications.

Affiliation: Clemson University

Title:  Application-Oriented Quantum Algorithm (AOQA) Design

Abstract: In this synthesis talk, we discuss how application and orientation towards use drives quantum optimization algorithm design. The underlying framework is comprised of two parallel abstraction pillars. The first is primarily technical: quantum computing abstraction layers, including underlying physics, hardware, software, and algorithms. The second is mixed: the application domain abstraction includes intersections between the technical and the physical infrastructure,  purpose, and policies of the domain, i.e., a work-domain hierarchy. To develop application-oriented quantum algorithms, there is necessarily an intersection between the two pillars. We identify mechanisms to navigate these differing abstractions both as individual researchers and as algorithm-focused members of collaborative teams. To illustrate these ideas, we present two approaches from quantum scheduling and location optimization projects. We emphasize how algorithmic design choices are a function of abstraction layers and researchers’ mental models. We discuss how to identify potential novel research directions based on a motivating application, engage with practitioners, and consider implications to build towards practical quantum advantage.

Bio: Dr. Emily Tucker is the Dean’s Assistant Professor of Industrial Engineering at Clemson University. Her research focuses on developing new stochastic and quantum optimization approaches to improve access to social good. Key application areas include the resilience of critical supply chains and health systems. Dr. Tucker has been recognized by IISE with the Dr. Hamed K. Eldin Outstanding Early Career IE in Academia Award as well as the Outstanding Teaching Award from the IISE Logistics and Supply Chain Division. In addition to scholarly impact, her applied research has led to invited opeds, including in the New York Times, and advising on national policy.

Affiliation: Rigetti

Title:  Navigating the Path to Quantum Utility in Optimization via Quantum Preconditioning

Abstract: State-of-the-art classical solvers set a formidable bar for demonstrating quantum utility in optimization. In this talk, we trace a cohesive trajectory of hybrid strategies aimed at breaching this threshold. We begin by establishing how shallow quantum circuits extract powerful problem correlations, grounded by large-scale benchmarking on superconducting hardware that rigorously defines the current gap to classical advantage. To systematically bridge this gap, we focus on novel paradigms designed to circumvent physical hardware constraints, such as noise and qubit counts. In particular, we highlight Quantum Preconditioning, an approach that utilizes quantum circuits to transform input problems, drastically accelerating the convergence of best-in-class classical heuristics. Evaluating these methods on problems ranging from theoretical spin glasses to real-world energy grid optimization, we identify the mechanisms driving hybrid speedups and outline practical pathways toward quantum utility.

Relevant works: PRA 109, 012429 (2024), PRApplied 23, 014045 (2025), PRApplied 24, 044013 (2025), arXiv:2603.09838.

Bio: Maxime Dupont leads quantum applications research at Rigetti. His work sits at the intersection of quantum computing, condensed matter theory, and computational physics. Before Rigetti, he was a postdoctoral researcher at UC Berkeley and Berkeley Lab. He earned his Ph.D. in theoretical physics from the University of Toulouse, France.

Affiliation: Quantum Artificial Intelligence Laboratory, NASA & USRA Research Institute for Advanced Computer Science

Title:  Evaluating QAOA expectation values can be as hard as counting optimal solutions

Abstract: Evaluating expectation values is a critical task for variational quantum eigensolvers, and for parameterized quantum circuits and other quantum algorithms more generally. We consider the well-studied case of the Quantum Approximate Optimization Algorithm (QAOA) for the MaxCut problem. Recent work of Wang et al. [arXiv:2511.20212] showed this task to be NP-hard in general for any QAOA depth p≥2, complementing past results showing efficiently computable formulas for p=1 with arbitrary problem graphs. We sharpen this dichotomy showing that for p≥2 exact or exponentially precise cost expectation value evaluation is #P-hard under deterministic polynomial-time Turing reductions. Hardness at p≥2 is shown to remain even for evaluating single pairwise correlators ⟨Z⊗Z⟩, as well as for highly restricted sets of algorithm parameters. Our proof refines the NP-hardness construction of Wang et al. that recovers the maximum cut value from the largest exponent of a QAOA Laurent polynomial, utilizing a distinct and simpler construction that extracts a value proportional to the total number of maximum cuts, in addition to the optimal cut value. Thus we show that the QAOA expectation value hardness transition from p=1 to p=2 is not only from tractability to optimization hardness, but to that of counting optimal solutions. As an application we show our results imply analogous hardness results for computing gradients and Hessians of QAOA circuits.

arxiv.org/abs/2608.11385 

Affiliation: Massachusetts Institute of Technology

Title:  Promise of Graph Sparsification and Decomposition for Noise Reduction in QAOA

Abstract: We develop new approximate compilation schemes that significantly reduce the expense of compiling the Quantum Approximate Optimization Algorithm (QAOA) for solving the Max-Cut problem. Our main focus is on compilation with trapped-ion simulators using Pauli-X operations and all-to-all Ising Hamiltonian HIsing evolution generated by Mølmer-Sørensen or optical dipole force interactions, though some of our results also apply to standard gate-based compilations. Our results are based on principles of graph sparsification and decomposition; the former reduces the number of edges in a graph while maintaining its cut structure, while the latter breaks a weighted graph into a small number of unweighted graphs. Though these techniques have been used as heuristics in various hybrid quantum algorithms, there have been no guarantees on their performance, to the best of our knowledge. This work provides the first provable guarantees using sparsification and decomposition to improve quantum noise resilience and reduce quantum circuit complexity. We further demonstrate that significant improvements to the approximation ratio are obtained using decomposition in simulated trapped-ion experiments with dephasing noise. We anticipate these approximate compilation techniques will be useful tools in a variety of future quantum computing experiments.

Bio: Swati Gupta is an Associate Professor and the Class of 1947 Career Development Professor at the MIT Sloan School of Management, in the Operations Research and Statistics Group. Her research is on the foundations of optimization and AI, and on how those foundations must change for modern computational and societal challenges — spanning algorithmic fairness, healthcare, hiring, energy, and quantum computing. In the quantum setting, her work has focused on quantum-classical hybrid methods, including warm-started QAOA for combinatorial problems such as Max-Cut. Prior to MIT, she held the Fouts Family Early Career Professorship as an Assistant Professor at the H. Milton Stewart School of Industrial and Systems Engineering at Georgia Tech from 2018–2023, where she served as lead of Ethical AI in the NSF AI Institute for Advances in Optimization from 2021–2023. She received a Ph.D. in Operations Research from MIT in 2017, following a joint Masters and B.Tech in Computer Science from IIT Delhi. Her work has been recognized by the 2023 NSF CAREER Award, the JP Morgan Early Career Faculty Recognition in 2021, and the NSF CISE Research Initiation Initiative Award in 2019.

Affiliation: LMU Munich 

Title:  Constraint-Preserving Quantum Optimization via Indicator Functions

Abstract: One of the biggest challenges in quantum optimization algorithms is the efficient incorporation of constraints. While straightforward penalty-based approaches effectively enlarge and roughen the search space, the standardly employed mixer-based approaches often require deep circuit constructions. This talk highlights a frequently dismissed alternative: Encoding constraints through an indicator function that maps all infeasible solutions to a single, most-expensive cost value [Phys. Rev. A 112, 062605]. For the QAOA, this can be accomplished by controlling the application of the cost unitary on an ancillary qubit, that encodes the feasibility of the current superposition of solutions, which in turn can be computed using subroutines such as the QPE. This approach preserves the modular structure of the QAOA and allows for the combination of different constraint-enforcing techniques such as XY-mixers for Hamming-weight constraints and indicator functions for inequality constraints.

Bio: Jonas Stein has a background in computer science and recently finished his PhD in the field of Quantum Applications at LMU Munich. In October, he will start a tenure-track position as assistant professor to co-lead the Quantum Applications and Research Laboratory at LMU Munich. His research focuses on the resource-efficient application and development of quantum algorithms in the domains of optimization and artificial intelligence.