arXiv subjects

Hiroki Hamaguchi

Publications and source records attributed to Hiroki Hamaguchi.

7 recordsLinked to original sources

arXiv2026

Adaptive Gradient-Based Methods for a Broader Class of Optimization Problems under Performative Prediction

We study optimization under performative prediction, where deploying a model affects the future data distribution. For this setting, several gradient-based approaches have been proposed. However, they typically assume specific data distributions or loss functions, which limit their practical applicability. To overcome these limitations, we propose a gradient-based optimization method with convergence guarantees under substantially weaker assumptions. Our method explicitly estimates the induced distribution shift through finite differences. It enables higher-dimensional optimization across broader classes of loss functions and data distributions. We also propose a practical variant that reduces the number of samples required. Numerical experiments demonstrate that our proposed algorithms converge faster and more consistently than existing ones.

math.OC↗

arXiv2026

Bounded-depth spacetime lattice surgery for resource-efficient fault-tolerant quantum computation

Fault-tolerant quantum computing based on lattice surgery requires place-and-route compilation with low spacetime overhead. Routing, in particular, faces a basic tension between suppressing path conflicts through greater spatial allocation and exploiting the time direction to realize ancilla-efficient spacetime routing. Existing approaches do not fully resolve this trade-off while retaining compatibility with inner factory layouts and termination guarantees. Here we introduce double-slice routing, a constant-depth spacetime-routing method that uses two consecutive time slices with a guarantee that its kink-parity correction terminates under both planar and stacked architectures. We numerically benchmark the resulting compiler on Hamiltonian-simulation workloads to show that double-slice routing reduces compilation cost by up to a factor of 2.4 over a single-slice baseline. Compared to projective routing, an existing method that allows an unbounded number of time slices per path, double-slice routing achieves smaller circuit volume with only a marginal execution-time penalty. Combined with a cultivation-compatible mapping optimization, the overall improvement reaches up to 7.5-fold over a naive single-slice compilation baseline. These results identify double-slice routing as a practically useful operating point in lattice-surgery compilation and show the substantial benefit in joint optimization of mapping and routing.

quant-ph↗

arXiv2026

Practical Regularized Quasi-Newton Methods with Inexact Function Values

Many practical optimization problems involve objective function values that are corrupted by unavoidable numerical errors. In smooth nonconvex optimization, quasi-Newton methods combined with line search are widely used due to their efficiency and scalability. These methods implicitly assume accurate function evaluations and thus may fail to converge in noisy settings. Developing fast and robust quasi-Newton methods for such scenarios is therefore crucial. To address this issue, we propose a noise-tolerant regularized quasi-Newton method equipped with a relaxed Armijo-type line search, designed to remain stable under inaccurate function evaluations. By combining a regularization parameter update rule inspired by Objective-Function-Free Optimization and the AdaGrad-Norm method, we establish a global convergence rate of $\mathcal{O}(1/\varepsilon^2)$ for reaching a first-order stationary point under the assumed error model. We performed extensive experiments on the CUTEst benchmark collection with artificially noisy objective function evaluations, as well as with low-precision floating-point arithmetic (64-, 32-, and 16-bit). The results demonstrate that the proposed method is substantially more robust than several existing methods, while maintaining competitive practical convergence speed and computational cost.

math.OC↗

arXiv2024

Initial Placement for Fruchterman--Reingold Force Model With Coordinate Newton Direction

The Fruchterman--Reingold (FR) force model is widely used in force-directed graph drawing, and multilevel approaches such as sfdp in Graphviz scale these methods effectively. A crucial step in multilevel schemes is refinement, which improves the graph layout, typically performed with the simulation-based algorithm. For this refinement, we can utilize optimization-based methods such as L-BFGS, or combine them with initial placement methods such as Simulated Annealing to achieve better layouts. However, they have several limitations, such as suffering from high per-iteration costs for large graphs or having difficulty with weighted and structurally complex graphs, leaving room for improvement. In this research, we propose a new initial placement based on stochastic coordinate descent to accelerate the optimization process. We first reformulate the problem as a discrete optimization problem using a hexagonal lattice and then iteratively update a randomly selected vertex along the coordinate Newton direction with low per-iteration costs. We demonstrate the effectiveness of our method through numerical experiments, showing that our initial placement leads to faster convergence and higher-quality layouts compared to naive optimization approaches. We also discuss applications of our method, such as drawing for Hooke--Coulomb and Eades force models.

cs.CG↗

arXiv2024

Handbook for Quantifying Robustness of Magic

The nonstabilizerness, or magic, is an essential quantum resource to perform universal quantum computation. Robustness of magic (RoM) in particular characterizes the degree of usefulness of a given quantum state for non-Clifford operation. While the mathematical formalism of RoM can be given in a concise manner, it is extremely challenging to determine the RoM in practice, since it involves superexponentially many pure stabilizer states. In this work, we present efficient novel algorithms to compute the RoM. The crucial technique is a subroutine that achieves the remarkable features in calculation of overlaps between pure stabilizer states: (i) the time complexity per each stabilizer is reduced exponentially, (ii) the space complexity is reduced superexponentially. Based on this subroutine, we present algorithms to compute the RoM for arbitrary states up to $n=7$ qubits on a laptop, while brute-force methods require a memory size of 86 TiB. As a byproduct, the proposed subroutine allows us to simulate the stabilizer fidelity up to $n=8$ qubits, for which naive methods require memory size of 86 PiB so that any state-of-the-art classical computer cannot execute the computation. We further propose novel algorithms that utilize the preknowledge on the structure of target quantum state such as the permutation symmetry of disentanglement, and numerically demonstrate our state-of-the-art results for copies of magic states and partially disentangled quantum states. The series of algorithms constitute a comprehensive ``handbook'' to scale up the computation of the RoM, and we envision that the proposed technique applies to the computation of other quantum resource measures as well.

quant-ph↗

arXiv2024

Sample Complexity of Low-rank Tensor Recovery from Uniformly Random Entries

We show that a generic tensor $T\in \mathbb{F}^{n\times n\times \dots\times n}$ of order $k$ and CP rank $d$ can be uniquely recovered from $n\log n+dn\log \log n +o(n\log \log n) $ uniformly random entries with high probability if $d$ and $k$ are constant and $\mathbb{F}\in \{\mathbb{R},\mathbb{C}\}$. The bound is tight up to the coefficient of the second leading term and improves on the existing $O(n^{\frac{k}{2}}{\rm polylog}(n))$ upper bound for order $k$ tensors. The bound is obtained by showing that the projection of the Segre variety to a random axis-parallel linear subspace preserves $d$-identifiability with high probability if the dimension of the subspace is $n\log n+dn\log \log n +o(n\log \log n) $ and $n$ is sufficiently large.

math.CO↗

arXiv2024

Faster computation of nonstabilizerness

The characterization of nonstabilizerness is fruitful due to its application in gate synthesis and classical simulation. In particular, the resource monotone called the stabilizer extent is a useful tool to estimate the simulation cost using rank-based simulators, one of the state-of-the-art simulators of Clifford+$T$ circuits. In this work, we propose faster numerical algorithms to compute the stabilizer extent. Our algorithm utilizes the Column Generation method, which iteratively updates the subset of pure stabilizer states used for calculation. This subset is selected based on the overlaps between all stabilizer states and a target state. In order to update the subset, we make use of a newly proposed subroutine for calculating the stabilizer fidelity that (i) achieves linear time complexity with respect to the number of stabilizer states, (ii) super-exponentially reduces the space complexity by in-place calculation, and (iii) prunes unnecessary states for the computation. As a result, our algorithm can compute the stabilizer fidelity and the stabilizer extent for Haar random pure states up to $n=9$ qubits, which naively requires a memory of 305 EiB. We further show that our algorithm runs faster when the target state vector is real. We prove that the problem size is reduced by $\mathcal{O}(2^n)$ compared to the general cases, which makes it computable for the case of $n=10$ qubits.

quant-ph↗