SQAI-NCTS Joint Workshop 2026

24 Aug 2026, 09:30 → 28 Aug 2026, 17:00 Asia/Tokyo

Fujiwara Hiroshi Hall, Kyoseikan (Hiyoshi Campus, Keio University, Yokohama, Japan)

Satoshi Morita (Keio University), Wei-Lin Tu (Keio University)

About

This workshop is the 3rd joint workshop by Center of Innovations for Sustainable Quantum AI (SQAI) and National Center for Theoretical Sciences Physics Division (NCTS Physics), focusing on a broad range of topics related to tensor networks and quantum computing. The topics also cover quantum algorithms, quantum machine learning, quantum embedding, condensed matter physics, and more.

Dates

August 24 - 28, 2026

Venue

Fujiwara Hiroshi Hall & Multipurpose Room 3, Kyoseikan
Hiyoshi Campus, Keio University, Yokohama, Japan
[Access and Floor Guide]

Registration fee

Free

Lecturers

 

• Philippe Corboz (Univ. of Amsterdam)
• Norbert Schuch (Univ. of Vienna)

Speakers

• Shinichiro Akiyama (Univ. Tsukuba)
• Fernando Brandão (AWS and Caltech)
• Giuseppe Carleo (EPFL)
• Juraj Hasik (Univ. Zurich)
• Hyun-Yong Lee (Korea Univ.)
• Seung-Sup Lee (Seoul Natl. Univ.)
• Dario Poletti (SUTD)
• Mingpu Qin (Shanghai Jiao Tong Univ.)
• Atsushi Ueda (Ghent Univ.)
• Nobuyuki Yoshioka (Univ. of Tokyo)

(in alphabetical order)

Organizers

Organizing Committee Chair

• Synge Todo (Univ. of Tokyo)
• Ying-Jer Kao (Natl. Taiwan Univ.)
• Pochung Chen (Natl. Tsing Hua Univ.)

Organizing Committee Members

• Chia-Min Chung (Natl. Yang Ming Chiao Tung Univ.)
• Ian McCulloch (Natl. Tsing Hua Univ.)
• Naoki Kawashima (ISSP, Univ. of Tokyo)
• Rico Pohle (Shizuoka Univ.)
• Satoshi Morita (Keio Univ.)
• Shunsuke Furukawa (Keio Univ.)
• Tsuyoshi Okubo (Niigata Univ.)
• Wei-Lin Tu (Keio Univ.)
• Yi-Ping Huang (Natl. Tsing Hua Univ.)
• Yusuke Nomura (Tohoku Univ.)

Supported by

• Center of Innovations for Sustainable Quantum AI (SQAI)
• National Center for Theoretical Sciences Physics Division (NCTS Physics)

Contact

Wei-Lin Tu
Graduate School of Science and Technology, Keio University
E-mail: weilintu@keio.jp

 

Satoshi Morita
Graduate School of Science and Technology, Keio University
E-mail: smorita@keio.jp

 

Note: If you receive any emails offering you hotels or accommodation for the event, e.g. from "Global Travel Experts" please ignore it - this is a well-known scam targeting registrants to Indico events.

Monday 24 August

Welcome Address

1 Opening speech

Speaker: Prof. Synge Todo (University of Tokyo)

Lectures

2 Lecture 1: Introduction to iPEPS

Infinite projected entangled-pair states (iPEPS) provide a state-of-the-art tool for studying strongly correlated systems in two dimensions directly in the thermodynamic limit. In the first lecture, I give a basic introduction to iPEPS, including standard contraction and optimization methods. The second lecture covers advanced iPEPS techniques and recent developments, including accelerated contractions, accurate evaluation of the energy variance with applications to the Shastry-Sutherland model, improved methods for computing excitation spectra based on the iPEPS excitation ansatz, and more.

Speaker: Prof. Philippe Corboz (Univ. of Amsterdam)

11:00 Coffer break

Lectures

3 Lecture 2: Advanced iPEPS techniques

Infinite projected entangled-pair states (iPEPS) provide a state-of-the-art tool for studying strongly correlated systems in two dimensions directly in the thermodynamic limit. In the first lecture, I give a basic introduction to iPEPS, including standard contraction and optimization methods. The second lecture covers advanced iPEPS techniques and recent developments, including accelerated contractions, accurate evaluation of the energy variance with applications to the Shastry-Sutherland model, improved methods for computing excitation spectra based on the iPEPS excitation ansatz, and more.

Speaker: Prof. Philippe Corboz (Univ. of Amsterdam)

12:20 Lunch

Invited talks

4 TBA

Speaker: Dr Atsushi Ueda (Ghent Univ.)

Contributed talks

5 Tensor Renormalization Group Approach to the 2D Classical ANNNI Model: Existence of the Floating Phase

We investigate the two-dimensional classical ANNNI model using the tensor renormalization group method. At finite next-nearest-neighbor interactions, the model is predicted to exhibit a critical phase known as the floating phase. In this study, we employ the bond-weighted tensor renormalization group method to investigate the floating phase. Using bond dimensions up to D=240, together with a finite-size scaling analysis, we explore the phase structure of the two-dimensional classical ANNNI model.

Speaker: Yuto Sugimoto (Tohoku University)

6 Copy-scarce learning of ground states

Probing ground-state properties is central to understanding quantum matter, yet conventional approaches usually require many independently prepared copies. This becomes prohibitive when ground-state preparation is costly and only a limited number of copies are available. Here we introduce a coherent readout protocol that processes given $P$ ground-state copies in parallel using controlled evolution under the system Hamiltonian and returns them nearly unchanged; the protocol simultaneously estimates $M$ possibly noncommuting observables to additive error $\varepsilon$, without requiring a coherent ground-state preparation circuit or its inverse. Its worst-case elapsed Hamiltonian evolution time is $\tilde{\mathcal{O}}({\Delta}^{-1} (\varepsilon^{-1}{\sqrt M}/P+1))$, where $\Delta$ is the spectral gap; structured observable sets further improve the $M$ dependence. We establish lower bounds of elapsed evolution time for arbitrary parallel protocols, even when the input copies may be consumed, proving simultaneous optimality in $M,\varepsilon,P$ and $\Delta$ up to logarithmic factors. Among its applications, the protocol performs full tomography of a $d$-dimensional ground state within trace-distance error $\eta$ using $\tilde{\mathcal{O}}(\Delta^{-1}(\eta^{-1}{d}/{P}+1))$ elapsed evolution time, with a nearly matching lower bound. These results establish the fundamental dynamical cost of learning ground states when state copies are scarce.

Speaker: Kaito Wada (International Center for Elementary Particle Physics, The University of Tokyo)

15:40 Coffee break

Invited talks

7 Walking Renormalization-Group Flow in the Dilute Baxter—Wu Model

Weak first-order transitions can exhibit extended pseudo-critical regimes that are notoriously difficult to distinguish from genuine criticality by conventional finite-size scaling. We show that the long-standing apparent criticality of the dilute Baxter—Wu model is instead a manifestation of an extremely slow renormalization-group flow due to a nearly marginal perturbation, ultimately leading to a weak first-order transition. Combining \rev{symmetry analysis} and conformal perturbation theory with tensor-network renormalization, we characterize the ``walking" mechanism by the deformation of the marginal operator spectrum and the beta function of an associated slowly running coupling through quartic order. Our diagnosis extends to the self-dual generalized Baxter—Wu model with distinct interactions on up- and down-pointing triangles, which, as it turns out, follows the same step-scaling function, revealing the local renormalization-group flow structure on both the marginally relevant and irrelevant sides across the four-state Potts point.

Speaker: Prof. Hyun-Yong Lee (Korea Univ.)

Tuesday 25 August

Invited talks

8 Tensor networks for quantum, quantum for tensor networks: two stories

Tensor network methods and quantum information science have enriched each other over the past few decades. In this talk, I will present two examples of this interplay from our recent works. First, in the direction of “tensor networks for quantum,” I will discuss the complexity of classically simulating a two-dimensional quantum sampling architecture in the presence of disorder. Using exact tensor-network contractions to evaluate output probabilities, we show that disorder in two-qubit interactions induces two crossovers, each undermining one of the two conjectured ingredients underlying sampling hardness. In the opposite direction, “quantum for tensor networks,” we develop a new algorithm for disentangling tensor trains using a restricted set of two-qubit Clifford gates. These disentangling transformations enable more efficient tensor cross interpolation at essentially no additional computational cost.

Speaker: Prof. Seung-Sup Lee (Seoul National University)

11:00 Coffee break

Contributed talks

9 Bootstrapping problem in tensor networks

Accurate contraction of tensor networks beyond one dimension is essential in various fields including quantum many-body physics. However, existing approaches typically rely on approximate contraction schemes and do not provide certified error bars. In this talk, we introduce an alternative perspective on tensor-network contraction problems via the numerical bootstrap framework. This technique casts the problem into a convex optimization problem, thereby yielding certified lower and upper bounds on expectation values of physical observables. As a proof-of-principle, we construct such constraints explicitly for translationally invariant matrix product states and demonstrate that our scheme can provide tight bounds on the contraction result. Our work suggests numerical bootstrap could be a possible way forward for the rigorous contraction of higher-dimensional tensor networks.

Speaker: Seishiro Ono (Institute for Solid State Physics, University of Tokyo)

10 Spectral Small-Incremental-Entangling and Tensor-Network Complexity

Understanding the complexity of quantum many-body states requires control not only over entanglement entropy but also over the distribution of entanglement across the Schmidt spectrum. The Small-Incremental-Entangling (SIE) theorem provides a remarkably general bound on the rate of change of bipartite von Neumann entanglement entropy, yet leaves this finer spectral structure unresolved. In this work, we introduce the Spectral-Entangling Strength and establish a spectral SIE theorem that bounds the rate of Rényi entanglement growth for all $\alpha \geq 1/2$. This implies a universal tail bound on the entanglement spectrum, with the threshold $\alpha = 1/2$ being optimal. This spectral control yields rigorous Schmidt-truncation bounds, thereby constraining the complexity of tensor networks. As applications, we establish a generalized entanglement area law along adiabatic paths beyond geometric locality and show that one-dimensional systems with long-range interactions admit polynomial-bond-dimension approximations for ground states, time-evolved states, and thermal states. Our results also provide an a priori precision guarantee for time-dependent density matrix renormalization group (tDMRG) simulations.

Speaker: Dr Donghoon Kim (RIKEN)

11 Isometrization of Tensor Network States via Gauge Propagation

We introduce a gauge-propagation approach for approximately converting generic tensor-network states into an isometric tensor-network state form with a prescribed orthogonality center. In one dimension, this propagation is exact because the non-isometric factor produced by a QR or singular-value decomposition is supported on a single virtual bond. In higher-dimensional networks, however, a local step can have several outgoing directions, and the residual factor is generally not separable into independent single-bond contributions.

We address this local obstruction by approximating a local tensor, or a contracted local cluster, by structured terms consisting of an isometric factor multiplied by a tensor product of output-leg factors. The isometric factor is retained at the current site or cluster, while the output-leg factors are absorbed into neighboring tensors along the propagation directions. This construction provides a local truncation criterion for gauge propagation and a practical route to refinement by increasing the number of retained terms or enlarging the local cluster.

Benchmarks on random tensors and on the loop-gas tensor representation of the Kitaev spin liquid show that this refinement reduces both local residuals and accumulated propagation errors. For the loop-gas tensor, two structured terms reduce the local residual to numerical precision, and enlarging the local object from 2-in-2-out to 4-in-2-out and 6-in-2-out clusters lowers both local truncation errors and accumulated errors in finite honeycomb gauge propagation.

These results identify propagation-compatible local decomposition as a useful building block for approximate isometrization and as a potential initializer or preconditioner for variational isoTNS algorithms.

Speaker: Zhiyu Jiang (The University of Osaka)

12:20 Lunch

Invited talks

12 Quantum resource in different perspectives and classically simulatable quantum many-body states

I will begin by examining different perspectives on what constitutes a quantum resource, specifically focusing on definitions that distinguish quantum systems from classical ones based on the criterion of classical simulability. I will then demonstrate how integrating various classically simulatable ingredients can lead to the development of more powerful classical algorithms for simulating quantum many-body systems.

Speaker: Prof. Mingpu Qin (Shanghai Jiao Tong University)

Contributed talks

13 Angular-time evolution and its correspondence to edge-spin dynamics in the Haldane chain

We discuss the angular-time evolution ---a parameter-time evolution generated by the entanglement Hamiltonian--- for the bipartitioned ground state of the S=1 bilinear–biquadratic chain under open boundary conditions. For the exactly sobvable AKLT chain, we first demonstrate the relationship between the angular-time evolution and edge-spin dynamics in a uniform magnetic field applied to the system part, using a gauge transformation of the matrix-product state. We then calculate angular-time spin correlation functions for the S=1 Heisenberg chain and extract the dominant oscillation mode originating from twofold-degenerate entanglement spectrum protected by the Z2×Z2 symmetry. Finally, we verify the correspondence between the edge-spin dynamics in a uniform magnetic field and this dominant angular-time mode, which may provide a new route to access the SPT entanglement of the Haldane phase through the observable edge-spin dynamics.

Speaker: Prof. Kouichi Okunishi (Department of Physics, Graduate School of Science, Osaka Metropolitan University)

14 Efficient Evaluation of Genuine Multipartite Negativity in Quantum Many-Body States

Genuine multipartite entanglement captures collective quantum correlations that cannot be described by convex mixtures of states separable across different bipartitions. Quantifying such correlations in quantum many-body states is essential for characterizing their irreducibly collective quantum structure and assessing their potential as multipartite quantum resources, yet remains computationally demanding: both the dimension of the reduced density matrix (RDM) and the number of bipartitions that must be considered grow exponentially with subsystem size. Here, by exploiting the bipartition-wise structure of the constraints in the underlying semidefinite program (SDP), we develop an efficient parallel optimization framework for evaluating genuine multipartite negativity (GMN) from many-body RDMs. Our approach enables GMN calculations for RDMs of up to 10–11 qubits, compared with a practical limit of around 6 qubits for standard SDP solvers. By extending the accessible subsystem size, our framework enables systematic mapping of GMN over a broader range of cluster sizes and geometries in quantum many-body systems, facilitating comparisons across physical regimes.

Speaker: Fengfeng Song (ISSP UTokyo)

15:40 Coffee break

Invited talks

15 TBA

Speaker: Prof. Dario Poletti (Singapore University of Technology and Design)

Wednesday 26 August

Lectures

16 Lecture 1: Mathematical structure of tensor networks

These lectures will provide an introduction into the mathematical theory of tensor networks, covering in particular how tensor networks can be used to construct solvable quantum many-body models, and how symmetries can be encoded in tensor networks, leading to the realization of symmetry-protected, symmetry-enriched, and topological phases of matter. The second lecture will moreover cover some recent mathematical results on tensor network models for quantum many-body systems.

Speaker: Prof. Nobert Schuch (Univ. of Vienna)

11:00 Coffee break

Lectures

17 Lecture 2: Mathematical structure of tensor networks

These lectures will provide an introduction into the mathematical theory of tensor networks, covering in particular how tensor networks can be used to construct solvable quantum many-body models, and how symmetries can be encoded in tensor networks, leading to the realization of symmetry-protected, symmetry-enriched, and topological phases of matter. The second lecture will moreover cover some recent mathematical results on tensor network models for quantum many-body systems.

Speaker: Prof. Nobert Schuch (Univ. of Vienna)

Poster: Poster Session with Light Refreshments

Poster number

18 Extracting Conformal Data from Critical 2D Classical Models using Tensor-Network-Based Finite-size Scaling

Critical 2D classical lattice models are analyzed through a method-agnostic finite-size-scaling framework built on tensor-network transfer matrices. The key idea is an explicit crossover length scale that separates the finite-size-scaling regime from the finite-entanglement-scaling regime induced by bond-dimension truncation. Working within this self-consistent finite-size window, the central charge, scaling dimensions, and conformal spins can be estimated and cross-validated across three independent tensor renormalization schemes — HOTRG, PTMRG, and CTRG — on the critical Ising and three-state clock models. Universal behavior emerges robustly below the crossover scale, with accurate conformal data extracted up to relatively high conformal levels. Moreover, a natural operational definition of entanglement scaling is provided by our analysis.

Speaker: Sing Hong Chan (National Tsing Hua University)

19 Telum.jl: A Julia-based non-Abelian tensor library with exact Clebsch–Gordan data

Telum.jl is a non-Abelian tensor network library written entirely in Julia. Its internal Clebsch–Gordan engine, LurCGT.jl, extends the approaches of QSpace and TensorKit.jl to compute symmetry-related quantities for SU(N), Sp(2N), SO(N), and G2 exactly using integer arithmetic and in a deterministic manner. Telum.jl is designed to handle tensors with arbitrary combination of these symmetries, with ready-to-use support for spinful fermionic spaces involving U(1), SU(2), and SU(N) charge, spin, and channel symmetries.
Telum.jl is designed for users to implement high-performance tensor network algorithms with simple and readable syntax. It provides an ITensor-style interface, together with utility methods for explicitly controlling contracted legs. On the performance side, Telum.jl implements an efficient SVD algorithm that avoids explicit leg fusion, as well as lazy evaluation for simple operations such as conjugation. In benchmark calculations performed so far, Telum.jl achieves CPU runtimes comparable to QSpace while reducing memory usage. Further optimizations, including lazy evaluation for tensor contractions and GPU support, are planned for future releases.

Speaker: Kiyeon Kim (Seoul National University)

20 Quantum algorithm for simulating metabolic networks

Biological systems, such as cellular metabolism, involve thousands of reactions that together determine how cells grow, respond to their environment, and produce energy. These are interconnected chemical reactions forming metabolic networks. Modeling and analyzing these systems require solving very large mathematical problems that can quickly become computationally prohibitive. To address this challenge, we present a quantum algorithm to analyze metabolic networks, focusing on flux balance analysis as a representative case. We use a quantum interior point method consisting of a quantum subroutine for matrix inversion. Specifically, we reformulate the metabolic optimization problem for efficient execution on a quantum computer using quantum singular value transformation, enabling a rapid solution of complex systems that arise in flux balance analysis. This quantum approach offers a potential computational advantage over classical interior point methods for large and well-conditioned networks. We demonstrate the practical applicability of our method with numerical simulations on the glycolysis and tricarboxylic acid (TCA) cycle network and show that the quantum solution converges to the correct biological objective.

Speaker: Ashish Joshi (Keio University)

21 Parallelization of isometric tensor networks

Isometric tensor networks provide globally optimal low-rank approximations without environment computation. However, this property is guaranteed only at the orthogonality center, making node-level parallelization difficult. In contrast, node-level parallelization has been developed based on Vidal canonical form [1,2], though the canonical conditions and the truncation optimality are gradually lost during variational optimization and time evolution. We propose a two-layer network architecture which combines the advantages of both representations. One layer strictly maintains an isometric structure, while the other use Vidal canonical form. This hybrid design aims to balance approximation accuracy and computational parallelism.

[1] G. Vidal, Phys. Rev. Lett. 91, 147902 (2003).
[2] R. -Y. Sun, T. Shirakawa, and S. Yunoki, Phys. Rev. B 110, 085149 (2024).

Speaker: Junya Yokokura (Department of Physics, Graduate School of Science, The University of Tokyo)

22 Nonorientable surfaces for non-Hermitian systems

The Klein bottle ratio provides universal values other than the central charge. Here, we extend the Klein bottle ratio and the $\mathrm{RP}^2$ ratio to non-Hermitian systems with periodic boundary conditions. Using the Yang-Lee model, we confirm the behavior of these ratios, as well as the generalized entanglement entropy. We also observe universal behaviors of these ratios for the non-Hermitian critical 5-state Potts model.

Speaker: Haruki Shimizu (ISSP, The university of Tokyo)

23 A conformal field theory approach to error bounds for localized virtual purification

Quantum error mitigation is a collection of techniques for reducing the effect of noise when estimating expectation values on noisy quantum computers. In particular, fully virtual purification (FVP) suppresses errors by preparing multiple copies of a noisy quantum state and cyclically permuting them, thereby amplifying the contribution of the dominant eigenstate that is least affected by noise. Although FVP achieves exponential error suppression with respect to the number of copies, it requires nonlocal operations across different copies as well as a large number of measurements, resulting in a substantial computational cost. To reduce this overhead, localized virtual purification (LVP) has recently been proposed, in which the cyclic permutation is applied only in the vicinity of the observable being measured [1]. While the expectation values obtained from FVP and LVP generally differ, it has been shown that, for systems with a unique gapped ground state, this difference decays exponentially with the distance (d(A,C)) between the support (A) of the observable and the region (C) where the cyclic permutation is not applied [1]. This result is based on a general argument using the Lieb–Robinson bound and therefore cannot be directly extended to gapless quantum critical systems.

In this work, we investigate the error of LVP relative to FVP in one-dimensional quantum many-body systems using conformal field theory (CFT). By extending the replica method originally developed for the analysis of entanglement entropy [2], we show that, for critical ground states, the error in the expectation values of subsystem Hamiltonians and two-point correlation functions exhibits a power-law decay as a function of the distance (d(A,C)). Furthermore, we demonstrate that, for finite-temperature Gibbs states of critical systems, the error decays exponentially with distance.

References

[1] H. Hakoshima, S. Endo, K. Yamamoto, Y. Matsuzaki, and N. Yoshioka, Phys. Rev. Lett. 133, 080601 (2024).

[2] P. Calabrese and J. Cardy, J. Stat. Mech. P06002 (2004).

Speaker: Mizuki Hamada (Keio University)

24 Information-Density Threshold for Configuration-Space Truncation in the Periodic XX Chain

Configuration-space truncation is a basic step in quantum-selected configuration interaction and related sample-based diagonalization methods, but the number of basis configurations required to retain a prescribed fraction of a many-body state is generally unknown. We study this intrinsic truncation problem for the fixed-particle-number ground state of the periodic XX, or free-fermion, chain. Its occupation-basis Born probabilities form a discrete circular log gas. Interpreting $Y(X)=-\log p(X)$ as an effective configuration energy, we derive exact expressions at arbitrary fixed filling for its mean density $h(\rho)$, equal to the Shannon participation-entropy density, and its leading fluctuation density $v(\rho)$, equal to the varentropy density. We prove that $\frac{Y}{L}$ becomes self-averaging, with $\operatorname{Var}(Y)=L v(\rho)+o(L)$, so that asymptotically all Born weight is concentrated in a microcanonical-like configuration-energy shell around $Y\simeq Lh(\rho)$. Consequently, optimal truncation to the (M) most probable configurations has a sharp threshold at the level of exponential rates: the retained probability tends to zero when $L^{-1}\log M<h(\rho)$, and to one when $L^{-1}\log M>h(\rho)$. Our results provide an analytic benchmark for ideal support selection and connect wave-function truncation with participation entropy, log-gas thermodynamics, and information-density concentration.

Speaker: Yuheng Sui (Keio University)

25 Symmetry-resolved quantic tensor-train approach to the hydrogen molecule

We develop a symmetry-resolved QTT framework for two-particle wavefunctions, using permutation parity to target symmetric or antisymmetric spatial states. Applied to H₂ within the Born–Oppenheimer approximation, QTT-DMRG accurately reproduces the ground singlet X 1Σg+ and lowest triplet b 3Σu+ states, with errors reaching 10^-4 Hartree. The second-order Rényi entropy approaches ln(2) at large internuclear separation for both states, revealing their distinct correlation behavior.

Speaker: Mr QianCan Chen (National Tsing Hua University)

26 Robust propagating in-gap modes due to spin-orbit domain walls in graphene and Chiral chains with two valleys and disorder of finite correlation length

Robust propagating in-gap modes due to spin-orbit domain walls in graphene

Recent success in making twisted multilayered graphene and transition metal dichalcogenides, exhibiting Moiré patterns and consequent domain structures on the nanoscale, raises questions about possible topological electronic states on domain walls (DWs) in presence of multiple induced couplings. We investigate the possible topological electron modes bound to DWs between several types of gapped domains in graphene with spin-orbit coupling, by using both a spectral flow theorem in the continuum theory and tight-binding lattice models.
Surprisingly, we find that a DW across which Valley-Zeeman spin-orbit coupling changes sign, in presence of any constant Rashba spin-orbit, hosts in-gap modes protected by time-reversal symmetry and the bulkgap, although the bulkgap is topologically trivial. The modes are robust even to lattice backscattering on a sharp DW profile, and are similar to those discovered on electric potential DWs in gated bilayer graphene.

Chiral chains with two valleys and disorder of finite correlation length

In one-dimensional chiral systems, electronic states at energy E = 0 evade localization and show a divergent density of states (DOS). For N coupled chains with zero-correlation-length disorder, this divergence persists only for odd N, while even N yields a vanishing DOS. We model N = 2 chiral chains using a thin spinless graphene nanotube with disordered Semenoff mass and Haldane coupling, introducing disorder with tunable correlation length. Because the two valleys at opposite momenta share correlated disorder, the system departs from analytical results assuming independent disorder channels. Numerical simulations show that the DOS remains suppressed for strongly coupled valleys and exhibits a nontrivial crossover as valleys decouple.

Speaker: Jean-Baptiste Touchais (National Cheng Kung University)

27 Disentangled Quantics Tensor Cross Interpolation

Quantics Tensor Cross Interpolation (QTCI) provides an efficient tensor-network representation of high-resolution functions by encoding physical coordinates into binary degrees of freedom and constructing a tensor train directly from sampled function values. Although QTCI is effective for many structured functions, its efficiency can deteriorate when the target function develops strong correlations across the binary variables, leading to large tensor-train bond dimensions. To address this limitation, we introduce a disentangling scheme based on invertible bit-string transformations generated from a GL(2,2)-based binary gate set. By reorganizing the binary coordinates before interpolation, this procedure seeks a representation in which the same sampled function admits a more disentangled tensor-train structure. We show that this disentangling approach improves the compression of otherwise high-rank functions at comparable accuracy, reducing the required bond dimensions and extending the practical applicability of QTCI to more challenging high-resolution functions.

Speaker: Mr Jeong-Hyeok Cha (Seoul National University)

28 An Origin of Kac-Type Long-Range Interactions

Long-range interacting quantum many-body systems have recently been realized in platforms such as trapped-ion systems and Rydberg atom arrays. In theoretical descriptions of long-range interactions, the Kac normalization factor is often introduced to keep the interaction energy per particle finite in the thermodynamic limit. In many models, however, this factor does not emerge naturally from the underlying interaction.

Frechette et al. showed that an elastic Ising model on a triangular lattice naturally generates a long-range effective interaction with a normalization factor proportional to the inverse number of particles. In this work, following the basic idea of Wagner and Horner, we treat the same model from the outset as a macroscopic two-dimensional isotropic elastic body rather than directly following the microscopic lattice deformation. For a finite system with free boundaries, the effective spin Hamiltonian takes the form

$$ H_{\mathrm{eff}} = -\frac{1}{\pi R^2} \sum_{n,m} \phi\!\left( \frac{\vec r_n}{R}, \frac{\vec r_m}{R} \right) S_nS_m , $$ where the inverse area $1/(\pi R^2)$ emerges naturally as the Kac normalization factor.

In this work, we show that a Kac-type long-range interaction and its normalization factor can arise from the macroscopic deformation of a finite elastic body, consistently with the microscopic triangular-lattice result. Although the present calculation is classical, the derivation is based entirely on macroscopic elasticity, suggesting that the same mechanism may also apply to quantum spin degrees of freedom.

Speaker: Yuji Okochi (Keio University)

29 Transfer-Matrix Calculations for Classical Spin Systems Using Clifford-Augmented Matrix Product States

Clifford-augmented matrix product states (CAMPS), which combine a Clifford circuit with a matrix product state (MPS), have recently been introduced as an efficient framework for quantum many-body simulations. In this work, we extend the CAMPS framework to transfer-matrix calculations for classical spin systems by exploiting the quantum–classical correspondence. We investigate the extent to which Clifford disentangling can reduce the bond dimension required for accurate computation of classical partition functions. We present benchmark results comparing the free-energy accuracy and the required bond dimensions of the CAMPS-based approach with those of conventional MPS-based transfer-matrix methods.

Speaker: Shoichiro Kado (Graduate School of Informatics, Kyoto university)

30 Characterization of the Loop O(n) Model Phase Diagram Based on the Global Properties of Phases via the Tensor Network Method

While the concept of the renormalization group (RG) is a highly powerful framework for the phase transitions, applying it analytically to specific systems is generally difficult. In recent years, the Tensor Renormalization Group (TRG), which utilizes Tensor Networks (TNs), has been developed as a method for performing RG calculations numerically with high precision. TRG can be easily executed even in regimes where conventional numerical techniques, such as Monte Carlo methods, encounter severe challenges. However, it suffers from the drawback that a systematic error evaluation method has not yet been established.

For numerical analyses using TNs, gauge-invariant partition function ratios have been proposed as a tool to evaluate errors via finite-size scaling. These ratios exhibit distinct values depending on the RG fixed point, and these values at a stable fixed point are believed to correspond to the number of thermodynamic states in that phase. Because partition function ratios have primarily been applied to spin systems, applying them to models described by different degrees of freedom is necessary to achieve a deeper understanding of their significance. Therefore, in this presentation, we focus on a class of systems known as classical loop O($n$) models, whose essential degrees of freedom can be considered as graphs on a lattice. The partition function of this model is expressed as a sum of weights over loop configurations defined on the lattice.

In this presentation, we investigate the cubic loop O($n$) model on square lattice, a variant of the classical loop O($n$) model that is closely related to the Ising model. For this system, we perform real-space renormalization using bond-weighted TRG—an improved variant of TRG—and compute the gauge-invariant partition function ratios. We reveal that as the system passes through two critical points, the calculated ratios change like $1 \to 2n \to n$. Furthermore, we demonstrate that these values do not necessarily align with the conventional interpretation of "the number of thermodynamical states."

Speaker: Ryoma Watanabe (ISSP, the University of Tokyo)

31 From coupled spin-1/2 chains to an anisotropic spin-1 chain: A numerical test of Schulz's framework

The spin-1 XXZ chain with uniaxial single-ion anisotropy exhibits a rich ground-state phase diagram, hosting topologically distinct Haldane and large-D phases alongside various magnetically ordered and gapless phases. In 1986, H. J. Schulz formulated an effective field theory for this system by mapping it onto coupled spin-1/2 chains (i.e., a ladder) and derived its phase diagram via bosonization and perturbative renormalization group (RG) techniques. While Schulz's results qualitatively agree with subsequent numerical studies of the spin-1 chain, it remains unclear how the phase structure of the anisotropic ladder evolves from the weak-interchain-coupling regime, where perturbative RG is most reliable, to the strong-coupling point that recovers the true spin-1 chain. In this study, we numerically investigate the evolution of the phase diagram across the full range of interchain coupling. We demonstrate that the qualitative features of the phase diagram are preserved throughout this evolution, confirming the robustness of Schulz's framework. However, we also highlight the critical role of competing magnetic phases unique to the ladder geometry.

Speaker: Rentaro Kuromi (Keio university)

32 Data-Driven Estimation of Conserved Quantities and Symmetry Structures in Quantum Many-Body Systems

The dynamics of quantum many-body systems are strongly constrained by the conserved quantities and symmetries of their Hamiltonians. Conserved quantities play a fundamental role in phenomena such as thermalization/nonthermalization, transport, and integrability, while the commutation relations among conserved quantities are closely related to continuous symmetries and the underlying algebraic structure of the system. In practice, however, information obtained from experiments and numerical simulations is generally limited to the time evolution of expectation values of a subset of observables, making it difficult to infer conserved quantities and the associated symmetry structure directly from the data.
Recent advances in machine learning have led to the development of powerful methods for analyzing physical systems. In quantum many-body physics, in particular, considerable progress has been made in Hamiltonian reconstruction from observational data [1]. By contrast, methods for inferring conserved quantities and symmetry structures from data have been developed mainly for classical dynamical systems [2-4]. Existing studies on quantum systems are limited to identifying conserved quantities with local support [5] and a general framework for inferring multiple conserved quantities together with their algebraic structure remains unavailable.
In this presentation, we propose a method for inferring multiple independent conserved quantities and the Lie algebra they generate from time-series data of expectation values in quantum many-body systems. Our approach combines Hamiltonian reconstruction with techniques for inferring conserved quantities and Lie algebraic structures originally developed for classical dynamical systems. Specifically, the proposed method consists of three steps: (i) reconstructing the Hamiltonian from time-series data of expectation values; (ii) parameterizing and optimizing conserved quantities so that they commute with the reconstructed Hamiltonian while remaining mutually independent; and (iii) computing the structure constants of the inferred conserved quantities to identify the underlying Lie algebra. Finally, we evaluate the proposed method using numerically generated time-series data of expectation values for both spin and fermionic systems.

[1] Rishabh Gupta et al., The Journal of Physical Chemistry A, 127 (2023)
[2] Artem Moskalev et al., arXiv:2210.04345
[3] Manu Bhat et al., arXiv:2504.10777
[4] Wanda Hou et al., arXiv:2412.14632
[5] T. E. O’Brien et al., Physical Review B, 94 (2016)

Speaker: Haruki Saito (Department of Physics, Graduate School of Science, the University of Tokyo)

33 Time-Uniform Error Bound for Temporal Coarse Graining in Markovian Open Quantum Systems

Quantum master equations are widely used to describe the dynamics of open quantum systems. Among them, the Redfield equation provides high accuracy but may generate unphysical time evolution because it does not preserve positivity. To overcome this issue, the Gorini–Kossakowski–Sudarshan–Lindblad (GKSL) equation, which generates completely positive and trace preserving (CPTP) dynamics, has been utilized in a wide range of fields, including quantum information, condensed matter physics, and quantum optics. However, rigorous error bounds for the deviation of the GKSL equation from the exact dynamics have been established only in the short-time regime.

In this work, we propose a unified framework, called temporal coarse graining, for describing the approximations used to derive GKSL equations from the Redfield equation. We show that several representative approximations proposed in previous studies are included in this framework. Furthermore, we derive an upper bound on the difference between the time evolutions generated by the GKSL and Redfield equations and prove that this error remains uniformly bounded in time. As a result, we provide a rigorous long-time accuracy guarantee for GKSL equations describing finite open quantum systems.

Speaker: Teruhiro Ikeuchi (Keio University)

34 Equivariant Tensor Renormalization on Finite-Cell Voltage-Graph Lifts

Higher-order tensor renormalization group (HOTRG) methods coarse-grain a tensor network by iterated blocking, implicitly assuming that the coarse network again admits a finite periodic description. We make this assumption explicit: representing a periodic tensor network as the lift of a finite-cell voltage graph over a group $\Lambda$, an expanding finite-index endomorphism of $\Lambda$ together with connected block templates (which we call a compatible equivariant block rule), represents one HOTRG step as an exact symbolically computable map between finite voltage data. We partially characterize when a compatible block rule can exist: it forces polynomial growth of the group (by Gromov's theorem these groups are virtually nilpotent). But we also note that polynomial growth is not sufficient, as the characteristically nilpotent Dixmier-Lister algebra has polynomial growth but no such endomorphism.

Our mathematical framework is able to represent several renormalizable geometries: triangular, square and hexagonal lattice. We validate exact partition-function agreement on these lattices. Moreover, lifts of wallpaper symmetry groups such as $p4,p4m,p6,p6m$ as well as the discrete Heisenberg group give rise to interesting new non-abelian geometries represented by the same framework. We show some (work-in-progress) tensor network calculations on each of these lattices to verify our HOTRG-inspired contraction procedure and investigate renormalizability of these geometries.

Speaker: Dr Sayan Mukherjee (The University of Tokyo)

35 Neural-network quantum states for solving few-body problems: application to Efimov physics

Neural-network quantum states have been developed as an efficient method for solving quantum many-body problems, not only in lattice systems but also in systems of particles in continuous space. Here, we apply this approach to strongly interacting few-body problems in continuous space at unitarity: the Efimov states and associated few-body bound states. We obtain the ground and first excited states of few bosons with a projection method, and a mass-imbalanced fermionic system consisting of two identical fermions and a third particle. The obtained energies of the ground and first excited states of these systems agree well with previously reported results. Furthermore, the proposed approach also reproduces the discrete scale invariance between the ground and first-excited states and the critical-mass behavior in mass-imbalanced fermionic systems. Our method can be straightforwardly applied to a broad class of strongly correlated few-body problems in continuous space.

Speaker: Ryuhei Furuta (University of Electro-Communications)

36 Non-Hermitian Magnon Dynamics in Floquet-Driven Antiferromagnets

Periodically driven magnetic systems provide a versatile platform for controlling collective spin excitations far from equilibrium. We theoretically study a two-dimensional antiferromagnet driven by in-plane polarized light, where the exchange interactions acquire Floquet modulations. The drive induces short-time amplification of selected magnon modes, which can be effectively described by a non-Hermitian magnon Hamiltonian. At longer times, nonlinear magnon interactions regulate the instability and lead to large-amplitude coherent oscillations. Our work identifies a mechanism by which a spatially local Floquet modulation of exchange interactions functions as a mode-selective magnon source. We further analyze the quantum geometry of the associated non-Hermitian magnon bands and discuss its implications for driven spin dynamics.

Speaker: Dr Jing Zhou (OIST)

37 Continuous-variable adiabatic quantum computation for combinatorial optimization via polyhedral domain decomposition

Adiabatic quantum computation is one approach to solving combinatorial optimization problems. In this talk, we propose a method that first transforms a combinatorial optimization problem into an equivalent continuous optimization problem via polyhedral domain decomposition and then performs adiabatic quantum computation. Since this formulation makes it easy to introduce quantum fluctuations tailored to the problem at hand, it is expected to search for optimal solutions more efficiently than conventional adiabatic quantum computation. We present a physical implementation based on a photonic quantum system and demonstrate through numerical simulations that optimal solutions are obtained with high probability for small but hard problem instances.

Speaker: Yuya Seki (Keio University)

38 Matrix Product Evolution: A Method for Simulating Quantum Circuits Using Tensor Networks

Efficient simulation of quantum circuits remains a central challenge in quantum information science. We introduce Matrix Product Evolution (MPE), a tensor-network framework that represents the evolution of each qubit as a tensor train organized along the temporal direction of a circuit. Temporal bond dimensions encode correlations accumulated during circuit evolution, and neighboring MPEs are contracted through a zip-up procedure with controlled bond truncation. We derive upper bounds on temporal bond dimensions for different classes of initial states and analyze the impact of post-selection on computational complexity. Numerical experiments on random quantum circuits and one-dimensional quantum Ising circuits demonstrate that the accuracy of MPE strongly depends on the underlying correlation structure. While MPE does not universally outperform conventional matrix product state approaches, it can achieve substantially higher accuracy for post-selected circuits by directly reducing effective temporal degrees of freedom. These results establish MPE as a complementary contraction strategy for tensor-network quantum-circuit simulation.

Speaker: Haruyuki Kawabe (BIPROGY Inc.)

39 Applying a QTT Space–Time Tensor-Network Solver to Gross–Pitaevskii Dynamics

This work applies a tensor-network space–time framework based on the Quantized Tensor Train (QTT) representation to the Gross–Pitaevskii equation (GPE), following the matrix-product-state space–time formulation introduced in Ref. [1]. The wavefunction is encoded as a Matrix Product State, while spatial differential operators, trapping potentials, temporal couplings, and nonlinear interaction terms are represented as Matrix Product Operators. By quantizing the spatial and temporal grids into binary tensor dimensions, the approach aims to reduce the storage and computational cost associated with large discretized wavefunctions.
The GPE is formulated as a space–time all-at-once system, in which the complete solution over the spatial and temporal domain is obtained simultaneously rather than through conventional sequential time stepping. The resulting tensor-network systems are solved using DMRG-based alternating optimization, while the nonlinear interaction is treated through self-consistent Picard iterations and relaxation. The implementation is progressively validated using harmonic-oscillator and GPE benchmark problems in one and two spatial dimensions. The framework is then applied to Bose–Einstein-condensate dynamics, including breathing-mode evolution motivated by previous QTT studies of the GPE [2]. These results provide a basis for investigating nonlinear quantum dynamics using compressed space–time tensor-network representations.
References
[1] R. D. Peddinti, S. Pisoni, N. Rapaka, M. K. Riahi, E. Tiunov, and L. Aolita, Quantum-inspired space-time PDE solver and dynamic mode decomposition, arXiv:2510.21767 (2025).
[2] Q.-C. Chen, I.-K. Liu, J.-W. Li, and C.-M. Chung, Solving the Gross-Pitaevskii Equation with Quantic Tensor Trains: Ground States and Nonlinear Dynamics, arXiv:2507.04279 (2026).

Speaker: Yen Chou (NYCU)

40 Classical Dimer Spin Liquids in Square and Square-Kagome Lattices

In this work, we introduce additional dimer degrees of freedom into classical spin models and investigate their thermodynamic behavior using Monte Carlo simulations. The dimers represent locally correlated spin pairs and are dynamically created, annihilated, and rearranged together with classical spin configurations. This semiclassical approach allows us to study the interplay between magnetic correlations and dimer formation beyond conventional classical spin Monte Carlo simulations. We apply this framework to the square and square-kagome lattices and examine how geometrical frustration affects the formation and spatial organization of dimers. Our results provide a simple classical framework for exploring dimer-dominated disordered states and their possible connection to spin-liquid-like behavior in frustrated magnets.

Speaker: Yusuke Kajiwara (Faculty of Science, Shizuoka University)

14:00 Free discussion

Thursday 27 August

Invited talks

41 TBA

Speaker: Prof. Giuseppe Carleo (EPFL)

11:00 Coffee break

Contributed talks

42 Thermalization from Real-Time Dynamics in the Two-Dimensional Hubbard Model

Thermalization in strongly correlated fermionic systems remains a central open problem in quantum many-body physics.
In this work, we investigate the real-time dynamics and the approach to thermalization in the two-dimensional Hubbard model, a paradigmatic framework for correlated electrons, relevant to high-temperature superconductivity and ultracold quantum simulation.
Focusing on the half-filled square lattice, we monitor the time evolution of the double occupancy following a quench in the on-site interaction $U$, and assess whether its long-time value is captured by a canonical thermal ensemble.
We employ time-dependent variational Monte Carlo methods combined with transformer-based Neural-Network Quantum States to accurately describe the nonequilibrium dynamics of fermions, especially for the behavior at long times, thereby accessing regimes that were previously inaccessible to numerical simulations.
Our results reveal two distinct dynamical behaviors: for weak to intermediate interactions, the long-time double occupancy agrees with the thermal prediction, consistent with ergodic relaxation; beyond a critical interaction $U_{C}$, the dynamics deviate markedly from the thermal expectation, revealing clear signatures of thermalization breaking or long-lived prethermal plateaux.
These results establish numerical simulation as a powerful tool to probe nonequilibrium quantum phenomena in correlated fermionic matter.

Speaker: Alessandro Sinibaldi (EPFL)

43 The Dual Power of Tensors: From Quantum Spin Liquids to High-Expressivity CNNs

High-dimensional correlations present a monumental challenge in both quantum physics and deep learning, typically requiring exponential computational resources or excessively deep architectures to resolve. This talk highlights low-rank tensor structures as a unifying solution to conquer complexity across these two distinct domains. First, we pivot to highly frustrated magnetism (arXiv:2606.31021), where symmetry-optimized infinite projected entangled-pair states (iPEPS) up to D=7 are used to solve the J₁-J₂ triangular Heisenberg model. Our simulation reveals that the 120 Neel state transits to a quantum spin liquid (QSL) phase at J₂/J₁ ≈ 0.08 and the QSL precludes Z₂ gapped nature, suggesting a U(1) Dirac spin liquid behavior. Secondly, we present the Tensor-Augmented CNN (TACNN; arXiv:2604.08072), which utilizes generic tensor kernels to map data into a virtual Hilbert space, showing very competitive accuracies in comparison with other advanced CNN architecture. Together, these works demonstrate the dual power of tensors as both an elegant variational tool for quantum ground states and an efficient architectural blueprint for deep learning.

Speaker: Wei-Lin Tu (Keio University)

44 Predictive Tensor-Network Tomography of Quantum Critical States from Reduced Density Matrices

A faithful reconstruction of a quantum many-body state enables the inference of physical properties not directly accessible from the input marginals. Achieving such predictive reconstruction is challenging because a tractable set of reduced density matrices does not generally determine the global state uniquely, and the many-body Hilbert space grows exponentially with system size. Here we develop a tensor-network tomography protocol that reconstructs a global matrix product state from a compact, physically designed set of reduced density matrices. Guided by the operator content of the critical theory, the selected subsystems combine nearest-neighbor spins, which probe local energy-like operators, with spatially separated spins, which constrain long-range spin correlations. The global state is optimized to minimize the quantum relative entropy between the input and reconstructed marginals for each selected subsystem. For critical Ising and three-state Potts chains up to around 64 sites, the protocol achieves global fidelities exceeding 99% and accurately reproduces universal long-distance correlation functions, substantially extending the accuracy and predictive scope demonstrated in earlier tensor-network tomography studies. Crucially, the reconstructed states recover nonlocal and universal properties not directly constrained by the tomographic input, including the universal scaling of the bipartite von Neumann entanglement entropy and the central charges, c=1/2 and c=4/5, of the two universality classes. Our results suggest a general design principle for efficient tomography of critical quantum states: select small marginals that simultaneously resolve the leading scaling dimensions via correlations.

Speaker: Yoshitomo Kamiya (Okinawa Institute of Science and Technology)

12:20 Lunch

Invited talks

45 TBA

Speaker: Dr Jurai Hasik (Univ. Zurich)

Contributed talks

46 Solving two-fermion problems using quantic tensor train

We present an approach to solve two-fermion problems using the Quantic Tensor Train (QTT) method. The QTT format allows us to represent high-resolution wavefunctions and operators with a memory cost that only scales logarithmically with the number of grid points. We demonstrate our method by showing our calculations on the ground state of a hydrogen molecule ($H_2$) and its quantum dynamics under an external electric field. Our results show that the QTT-based solver achieves controlled accuracy and successfully captures particle correlation effects.

Speaker: Chia-Min Chung (NYCU)

47 Practical efficiency of Clifford-circuit-augmented MPS beyond 1D open lattice

The matrix product state (MPS) can efficiently approximate low-entanglement states in 1D quantum many-body systems. However, applications to the gapless phase, periodic boundary case, and higher-dimensional systems are limited due to rapid growth of entanglement. Recently, Clifford-circuit-augmented MPS has been proposed as a hybrid ansatz of the Clifford circuit and MPS. The Clifford circuit is easily simulable on classical computers and plays the role of a disentangler. Thus, the CAMPS can describe strongly entangled states with smaller bond dimensions than the ordinary MPS. In this work, we developed a library of the CAMPS version of the density-matrix renormalization group (DMRG) and applied it to various models. We show that CAMPS can be a suitable ansatz for systems with periodic boundary conditions or in two dimensions.

Speaker: Akira Matsumoto (Osaka Metropolitan University)

15:40 Coffee break

Invited talks

Convener: Satoshi Morita (Keio University)

48 Recent progress in tensor networks for lattice field theories

Tensor networks offer a novel way to investigate lattice field theories. The community has been making various efforts toward their future application to QCD at finite temperature and finite density. In this talk, I will explain our recent studies, including methods for finite-temperature lattice gauge theories, Grassmann tensor networks, the use of symmetry-twisted partition functions, and QCD in the strong-coupling limit.

Speaker: Dr Shinichiro Akiyama (University of Tsukuba)

Friday 28 August

Invited talks

49 TBA

Speaker: Prof. Fernando Brandão (AWS and Caltech)

11:00 Coffee break

Contributed talks

50 Single-copy binary composite quantum hypothesis testing for qubit states under a minimax Bayesian error criterion

Hypothesis testing is a fundamental framework for making decisions between competing explanations of a system on the basis of observed data. In quantum information science, it underlies tasks such as signal detection, quantum sensing, communication, and the verification of quantum devices, where one must infer which quantum state or physical process generated the observed measurement outcomes. Standard quantum hypothesis testing assumes that each hypothesis specifies a single quantum state. In many realistic situations, however, unknown parameters, calibration errors, or incomplete prior information mean that the state under each hypothesis is only known to belong to a prescribed set. Composite quantum hypothesis testing addresses this uncertainty by seeking a state-independent measurement that performs optimally over all states consistent with the competing hypotheses.

We study single-copy binary composite quantum hypothesis testing for qubit states under a minimax Bayesian error criterion. The null hypothesis consists of a fixed pure state, while the alternative hypothesis is a continuous family of pure states whose Bloch vectors form an arc on a great circle. By exploiting the linearity of the error probability in the space of quantum states, it follows from Fang and Hayashi (2025) that replacing the original nonconvex alternative set by its convex hull leaves not only the minimum worst-case error probability but also the set of minimax-optimal POVMs unchanged. The optimization is then reduced to a geometric problem involving the convex hull of an affine image of the Bloch vector arc.

We derive an explicit characterization of the least favorable state and the optimal binary measurement for arbitrary prior probabilities and arc endpoints. Depending on the geometry and the prior imbalance, the least favorable state is either an endpoint pure state or a mixed state formed from the two endpoint states. Accordingly, the minimax-optimal measurement is either the Helstrom measurement for the null state and an endpoint state, the Helstrom measurement for the null state and a least favorable endpoint mixture, or a trivial decision rule that always selects the more probable hypothesis. We also identify the boundary cases in which the weighted state difference becomes singular and the optimal effect operator is nonunique. These results provide a complete Bloch-sphere interpretation of single-copy composite qubit discrimination and explicitly connect geometric convex optimization with general minimax results for composite quantum hypothesis testing.

Speaker: Shimpei Yamaguchi (Keio University)

51 Cross-Platform Quantum and Quantum-Inspired Optimization for Width-Aware Ecological Corridor Design

Under increasing environmental pressures, ecological corridor planning is vital for maintaining habitat connectivity. However, traditional approaches such as Minimum Cumulative Resistance (MCR) often generate singular, line-like paths that are highly vulnerable to local disruptions. In this study, we reformulate corridor design as a Quadratic Unconstrained Binary Optimization (QUBO) problem, explicitly incorporating volumetric awareness to enhance network resilience and obstacle avoidance. For large-scale scenarios, the proposed formulation is evaluated using the Compal GPU Annealer (CGA), Simulated Annealing (SA), and the D-Wave Quantum Annealer, while the simulator-based QAOA implementations using IBM Qiskit and NVIDIA CUDA-Q are employed for small-scale microscopic validation. Experimental results demonstrate that the proposed QUBO-based framework generates substantially wider and more spatially distributed ecological corridors than traditional methods while maintaining competitive ecological resistance. Furthermore, the proposed Corridor Efficiency (CE) and Robustness Gain Ratio (RGR) complement conventional resistance-based metrics by providing a more comprehensive evaluation of corridor quality. Ultimately, evaluations across these diverse solvers demonstrate that quantum and quantum-inspired annealing platforms successfully leverage the proposed QUBO model to achieve a highly favorable trade-off between ecological resistance and spatial redundancy.

Speaker: Chia-Ho Ou (Graduate School of Information Sciences, Tohoku University, Sendai, Japan)

52 Fractional-QUBO Optimization for Global and Nested SNP–Environment Barcode Discovery in Oral Cancer and OPMD

High-order combinations of genetic and environmental factors may reveal risk-associated subgroups in oral malignant disorders, but their exhaustive evaluation grows combinatorially, while heuristic methods provide no guarantee of global optimality. We formulate corrected-odds-ratio barcode discovery as a feature-level fractional quadratic unconstrained binary optimization (QUBO) problem. Binary variables encode selected factor states, profile-match auxiliary variables identify participants satisfying the complete barcode, and exactly-k and one-state-per-factor constraints enforce biologically valid solutions. The Haldane–Anscombe corrected odds ratio is optimized through Dinkelbach iterations.

The framework was evaluated using a public Taiwanese cohort of 576 participants, comprising 242 patients with oral or pharyngeal cancer, 70 patients with oral potentially malignant disorders (OPMD), and 264 controls. The search space included seven CYP26 single-nucleotide polymorphisms and seven demographic or environmental factors. For barcode sizes k=2,…,7, global-best and sequential nested-best solutions were obtained for cancer versus control, OPMD versus control, and cancer versus OPMD. Exhaustive enumeration, greedy search, binary particle swarm optimization, and simulated annealing were used for validation.

Across 36 global and nested scenarios, the minimum-energy QUBO solutions exactly matched the exhaustive corrected-OR optima. The generated models contained up to 2,119 binary variables and 55,613 nonzero quadratic coefficients. Global and nested paths coincided for cancer-versus-control and OPMD-versus-control, but diverged from k=3 in the cross-sectional cancer-versus-OPMD analysis. In this contrast, the global corrected OR reached 15.81 for k=4–6, with 24 cancer and no OPMD matches, whereas the nested path remained at 6.37 through k=5. This demonstrates that medically interpretable nested extensions do not necessarily preserve global optimality. Across the global problems, mean exact-optimum hit rates were 100%, 66.7%, 84.4%, and 95.6% for exhaustive enumeration, greedy search, binary particle swarm optimization, and simulated annealing, respectively.

At the present 14-factor scale, exhaustive enumeration remained the fastest exact reference. The results therefore establish an exactly validated real-data proof of concept for fractional-QUBO SNP–environment barcode optimization, rather than evidence of quantum advantage, and provide a foundation for subsequent large-scale digital- and quantum-annealing studies.

Keywords: fractional QUBO; SNP–environment barcode; corrected odds ratio; oral cancer; OPMD; combinatorial optimization

Speaker: Ms Shih-Han Huang (National Pingtung University)

12:20 Lunch

Invited talks

53 Gates and Observable Estimation in Early Fault-Tolerant Quantum Computing

Recent experimental advances in quantum error correction have pushed the field beyond the break-even point, offering a credible path toward early fault-tolerant quantum computing (FTQC) with logical qubits. In this emerging regime, two challenges become central: low-cost implementation of non-Clifford logical gates and full usage of syndrome information.
In the first part of this talk, we discuss approaches to realizing non-Clifford operations using weakly transversal gates, which provide a resource-efficient pathway compatible with near-term fault-tolerant architectures [1]. These constructions highlight how relaxed transversality conditions can circumvent traditional constraints with partial fault tolerance.
In the second part, we turn to the problem of observable estimation, focusing on protocols that leverage syndrome information. We present recent results demonstrating a separation between classical and quantum measurement strategies in this setting, emphasizing how error-correction data can be repurposed to enhance estimation performance [2].

[1] NY et al., arXiv:2510.08290
[2] Tsubouchi, Kwon, Jiang, NY, arXiv:2603.05145

Speaker: Prof. Nobuyuki Yoshioka (Univ. of Tokyo)

Contributed talks

54 Tree tensor network analysis of ancilla-assisted passive linear optical Bell-state discrimination

Bell-state measurements are central to photonic quantum information with applications in quantum teleportation, entanglement swapping, quantum repeaters, dense coding, and fusion-based photonic quantum computation. A passive linear optical interferometer with vacuum auxiliary modes and photon number resolving detectors can unambiguously identify at most two of the four equiprobable Bell states with an average success probability that cannot exceed 1/2. Ancillary photons provide a direct route beyond the one-half limit while retaining passive linear optics.

For a fixed circuit, the computational task is to determine which photon-counting patterns occur for exactly one of the four Bell inputs and to sum their probabilities. Individual passive-linear-optical amplitudes can be evaluated from matrix permanents, but unambiguous discrimination requires a simultaneous four-label classification measurement patterns. At the largest size considered here, M = 32 modes and Q = 30 photons give approximately 2.33 × 10^17 ideal full detector patterns. This rules out naive enumeration of the complete outcome space, but it does not imply that all structured evaluations are intractable.

In this research we sucessfully adapt a tree tensor network (TTN), whose bipartitions follow the recursive optical circuit of beam splitters, to ancilla-assisted linear-optical Bell-state discrimination. Photon-number U(1) symmetry resolves each virtual bond into fixed-charge sectors that significantly reduces complexity of computations and allows for finding probabilities of unambiguous state discrimination with no other than machine precision related approximations. To our knowledge, this is the first treatment to combine a U(1)-resolved TTN with simultaneous four-label detector-support classification for ancilla-assisted passive-linear-optical Bell-state discrimination.

The method reproduces known probabilities of Bell-state discrimination and additionally provides photon number resolving detector patterns uniquelly associated with each Bell state in terms of photon number prefix cylinders, which avoids enumeration of all possible patterns. Scalling is discussed.

Speaker: Wojciech Roga (Keio University)

55 Heuristic Schedule of Constrained Multi-Stage Quantum Walk for Combinatorial Optimization Problems with Equality Constraints

Constrained combinatorial optimization problems (CCOPs) arise in a wide range of practical applications. Quantum optimization methods have attracted considerable attention as heuristic approaches to solving these problems. Because current quantum hardware is limited by short coherence times and gate errors, optimization methods based on relatively short quantum evolutions are desirable.
Multi-stage quantum walks (MSQWs) perform optimization through a sequence of quantum evolutions under Hamiltonians with different parameters. A previous study proposed an efficient heuristic method for determining the MSQW parameters and evaluated it on Ising spin-glass instances [1]. That method targets unconstrained Ising problems, and its applicability to CCOPs has not yet been established.
In this study, we extend the heuristic parameter-setting method to CCOPs. We replace the transverse-field driver with an XY driver that preserves the Hamming-weight constraint, so that dynamics initialized in the feasible subspace remain confined to it. We apply the extended MSQW to CCOPs with equality constraints and evaluate its performance by state-vector simulation for systems of up to 12 qubits. Our results show that the extended heuristic determines effective parameters for CCOPs. Compared with a penalty-based formulation solved within the same MSQW framework, the constraint-preserving MSQW achieves higher success probabilities over the instances considered.
[1] A. Hopkins and V. Kendon. " Heuristics for multi-stage quantum walks to find Ising ground states," arXiv preprint arXiv:2511.01312 (2025).

Speaker: Tomohiro Hattori (Keio University)

Closing remark