Tootfinder

Adaptive Memory Momentum via a Model-Based Framework for Deep Learning Optimization
Kristi Topollai, Anna Choromanska
https://arxiv.org/abs/2510.04988 https://

Adaptive Memory Momentum via a Model-Based Framework for Deep Learning Optimization
The vast majority of modern deep learning models are trained with momentum-based first-order optimizers. The momentum term governs the optimizer's memory by determining how much each past gradient contributes to the current convergence direction. Fundamental momentum methods, such as Nesterov Accelerated Gradient and the Heavy Ball method, as well as more recent optimizers such as AdamW and Lion, all rely on the momentum coefficient that is customarily set to $β= 0.9$ and kept constant during …

Optimizing VQE Ansatz for Studying Tight-Binding Models with \textit{sd}-Interaction and On-Site Coulomb Repulsion
Oleg Udalov
https://arxiv.org/abs/2510.08864 https://

Optimizing VQE Ansatz for Studying Tight-Binding Models with \textit{sd}-Interaction and On-Site Coulomb Repulsion
The VQE algorithm is applied to the problem of finding the ground state of a lattice model with on-site Coulomb repulsion, nearest-neighbor hopping, and on-site sd-interaction. We compare the performance of several ansatze, including cluster and generic forms. Several modifications of the standard cluster ansatz implementation are proposed, which significantly reduce the number of two-qubit gates. Different classical optimizers are employed within the VQE algorithm. The performance of the algor…

Openness of shape optimizers in higher-order and non-scalar problems
Rupert L. Frank
https://arxiv.org/abs/2509.22487 https://arxiv.org/pdf/2509.22487

Openness of shape optimizers in higher-order and non-scalar problems
We consider the first eigenvalues of the polyharmonic, Lamé and Stokes operators with Dirichlet boundary conditions on sets of given finite measure. It is shown that a quasi-open set for which this eigenvalue is minimal is open. This removes dimensional restrictions in earlier works. We use Campanato theory, which works well in the present higher order or non-scalar setting.

Statistical Benchmarking of Optimization Methods for Variational Quantum Eigensolver under Quantum Noise
Silvie Ill\'esov\'a, Tom\'a\v{s} Bezd\v{e}k, Vojt\v{e}ch Nov\'ak, Bruno Senjean, Martin Beseda
https://arxiv.org/abs/2510.08727

Statistical Benchmarking of Optimization Methods for Variational Quantum Eigensolver under Quantum Noise
This work investigates the performance of numerical optimization algorithms applied to the State-Averaged Orbital-Optimized Variational Quantum Eigensolver for the H2 molecule under various quantum noise conditions. The goal is to assess the stability, accuracy, and computational efficiency of commonly used gradient-based, gradient-free, and global optimization strategies within the Noisy Intermediate-Scale Quantum regime. We systematically compare six representative optimizers, BFGS, SLSQP, Ne…

Efficient Differentiable Contact Model with Long-range Influence
Xiaohan Ye, Kui Wu, Zherong Pan, Taku Komura
https://arxiv.org/abs/2509.20917 https://arxi…

Efficient Differentiable Contact Model with Long-range Influence
With the maturation of differentiable physics, its role in various downstream applications: such as model predictive control, robotic design optimization, and neural PDE solvers, has become increasingly important. However, the derivative information provided by differentiable simulators can exhibit abrupt changes or vanish altogether, impeding the convergence of gradient-based optimizers. In this work, we demonstrate that such erratic gradient behavior is closely tied to the design of contact m…

Anisotropic fractional area measures
Xiaxing Cai
https://arxiv.org/abs/2510.05279 https://arxiv.org/pdf/2510.05279

Anisotropic fractional area measures
The anisotropic $s$-fractional area measures are introduced as the first variation of the anisotropic fractional $s$-perimeter $P_s(K,L)$, with $L$ an origin symmetric convex body and $s\in(0,1)$. As $s\rightarrow 1^-$, the anisotropic $s$-fractional area measure converges to the mixed area measure of $K$ and the moment body of $L$. The Minkowski problem of these measures are solved. Finally, a necessary condition for the convexity of optimizers in the anisotropic fractional isoperimetric inequ…

NeuSO: Neural Optimizer for Subgraph Queries
Linglin Yang, Lei Zou, Chunshan Zhao
https://arxiv.org/abs/2509.23775 https://arxiv.org/pdf/2509.23775

NeuSO: Neural Optimizer for Subgraph Queries
Subgraph query is a critical task in graph analysis with a wide range of applications across various domains. Most existing methods rely on heuristic vertex matching orderings, which may significantly degrade enumeration performance for certain queries. While learning-based optimizers have recently gained attention in the context of relational databases, they cannot be directly applied to subgraph queries due to the heterogeneous and schema-flexible nature of graph data, as well as the large nu…

Drop-Muon: Update Less, Converge Faster
Kaja Gruntkowska, Yassine Maziane, Zheng Qu, Peter Richt\'arik
https://arxiv.org/abs/2510.02239 https://arxiv.o…

Drop-Muon: Update Less, Converge Faster
Conventional wisdom in deep learning optimization dictates updating all layers at every step-a principle followed by all recent state-of-the-art optimizers such as Muon. In this work, we challenge this assumption, showing that full-network updates can be fundamentally suboptimal, both in theory and in practice. We introduce a non-Euclidean Randomized Progressive Training method-Drop-Muon-a simple yet powerful framework that updates only a subset of layers per step according to a randomized sche…

Optimizers of the Finite-Rank Hardy-Lieb-Thirring Inequality for Hardy-Schr\"odinger Operator
Bin Chen, Yujin Guo, Shuang Wu
https://arxiv.org/abs/2509.17307 https://

Optimizers of the Finite-Rank Hardy-Lieb-Thirring Inequality for Hardy-Schrödinger Operator
We study the following finite-rank Hardy-Lieb-Thirring inequality of Hardy-Schrödinger operator: \begin{equation*} \sum_{i=1}^N\left|λ_i\Big(-Δ-\frac{c}{|x|^2}-V\Big)\right|^s\leq C_{s,d}^{(N)}\int_{\mathbb R^d}V_+^{s+\frac d2}dx, \end{equation*} where $N\in\mathbb N^+$, $d\geq3$, $00$ is the best constant of Hardy's inequality, and $V\in L^{s+\frac d2}(\mathbb R^d)$ holds for $s>0$. Here $λ_i\big(-Δ-{c}{|x|^{-2}}-V\big)$…

Hybrid Sequential Quantum Computing
Pranav Chandarana, Sebasti\'an V. Romero, Alejandro Gomez Cadavid, Anton Simen, Enrique Solano, Narendra N. Hegade
https://arxiv.org/abs/2510.05851

Hybrid Sequential Quantum Computing
We introduce hybrid sequential quantum computing (HSQC), a paradigm for combinatorial optimization that systematically integrates classical and quantum methods within a structured, stage-wise workflow. HSQC may involve an arbitrary sequence of classical and quantum processes, as long as the global result outperforms the standalone components. Our testbed begins with classical optimizers to explore the solution landscape, followed by quantum optimization to refine candidate solutions, and conclu…

AERO-MPPI: Anchor-Guided Ensemble Trajectory Optimization for Agile Mapless Drone Navigation
Xin Chen, Rui Huang, Longbin Tang, Lin Zhao
https://arxiv.org/abs/2509.17340 https:/…

AERO-MPPI: Anchor-Guided Ensemble Trajectory Optimization for Agile Mapless Drone Navigation
Agile mapless navigation in cluttered 3D environments poses significant challenges for autonomous drones. Conventional mapping-planning-control pipelines incur high computational cost and propagate estimation errors. We present AERO-MPPI, a fully GPU-accelerated framework that unifies perception and planning through an anchor-guided ensemble of Model Predictive Path Integral (MPPI) optimizers. Specifically, we design a multi-resolution LiDAR point-cloud representation that rapidly extracts spat…

Replaced article(s) found for cs.LG. https://arxiv.org/list/cs.LG/new
[4/10]:
- A geometric framework for momentum-based optimizers for low-rank training
Steffen Schotth\"ofer, Timon Klein, Jonas Kusch