Maxim V. Rakhuba

42 papers A* 5A 5Journal 30Unranked 2

Year	Rank	Type	Title / Venue / Authors
2026	J	`jnl`	Optimization on the extended tensor-train manifold with shared factors. Comput. Appl. Math. Alexander Molozhavenko, Maxim V. Rakhuba
2025	J	`jnl`	Accelerating Newton-Schulz Iteration for Orthogonalization via Chebyshev-type Polynomials. CoRR Ekaterina Grishina, Matvey Smirnov, Maxim V. Rakhuba
2025	J	`jnl`	DyKAF: Dynamical Kronecker Approximation of the Fisher Information Matrix for Gradient Preconditioning. CoRR Nikolay Yudin, Ekaterina Grishina, Andrey Veprikov, Alexandr Beznosikov, Maxim V. Rakhuba
2025	A	`conf`	Knowledge Graph Completion with Mixed Geometry Tensor Factorization. AISTATS Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2025	J	`jnl`	Knowledge Graph Completion with Mixed Geometry Tensor Factorization. CoRR Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2025	A	`conf`	Leveraging Geometric Insights in Hyperbolic Triplet Loss for Improved Recommendations. RecSys Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2025	J	`jnl`	Leveraging Geometric Insights in Hyperbolic Triplet Loss for Improved Recommendations. CoRR Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2025	J	`jnl`	Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation. CoRR Askar Tsyganov, Evgeny Frolov, Sergey Samsonov, Maxim V. Rakhuba
2025	J	`jnl`	On the Upper Bounds for the Matrix Spectral Norm. CoRR Alexey Naumov, Maxim V. Rakhuba, Denis Ryapolov, Sergey Samsonov
2025	J	`jnl`	Optimization on the Extended Tensor-Train Manifold with Shared Factors. CoRR Alexander Molozhavenko, Maxim V. Rakhuba
2025	J	`jnl`	Pay Attention to Attention Distribution: A New Local Lipschitz Bound for Transformers. CoRR Nikolay Yudin, Alexander Gaponov, Sergei Kudriashov, Maxim V. Rakhuba
2025	—	`conf`	ProcrustesGPT: Compressing LLMs with Structured Matrices and Orthogonal Transformations. ACL (Findings) Ekaterina Grishina, Mikhail Gorbunov, Maxim V. Rakhuba
2025	J	`jnl`	ProcrustesGPT: Compressing LLMs with Structured Matrices and Orthogonal Transformations. CoRR Ekaterina Grishina, Mikhail Gorbunov, Maxim V. Rakhuba
2025	J	`jnl`	RiemannLoRA: A Unified Riemannian Framework for Ambiguity-Free LoRA Optimization. CoRR Vladimir Bogachev, Vladimir Aletov, Alexander Molozhavenko, Denis Bobkov, Vera Soboleva, Aibek Alanov, Maxim V. Rakhuba
2025	A	`conf`	Ultra Fast Warm Start Solution for Graph Recommendations. CIKM Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2025	J	`jnl`	Ultra Fast Warm Start Solution for Graph Recommendations. CoRR Viacheslav Yusupov, Maxim V. Rakhuba, Evgeny Frolov
2024	A*	`conf`	Dimension-free Structured Covariance Estimation. COLT Nikita Puchkin, Maxim V. Rakhuba
2024	A*	`conf`	Group and Shuffle: Efficient Structured Orthogonal Parametrization. NeurIPS Mikhail Gorbunov, Nikolay Yudin, Vera Soboleva, Aibek Alanov, Alexey Naumov, Maxim V. Rakhuba
2024	J	`jnl`	Group and Shuffle: Efficient Structured Orthogonal Parametrization. CoRR Mikhail Gorbunov, Nikolay Yudin, Vera Soboleva, Aibek Alanov, Alexey Naumov, Maxim V. Rakhuba
2024	—	`conf`	Tight and Efficient Upper Bound on Spectral Norm of Convolutional Layers. ECCV (89) Ekaterina Grishina, Mikhail Gorbunov, Maxim V. Rakhuba
2024	J	`jnl`	Tight and Efficient Upper Bound on Spectral Norm of Convolutional Layers. CoRR Ekaterina Grishina, Mikhail Gorbunov, Maxim V. Rakhuba
2024	A	`conf`	Training a Tucker Model With Shared Factors: a Riemannian Optimization Approach. AISTATS Ivan Peshekhonov, Aleksey Arzhantsev, Maxim V. Rakhuba
2023	J	`jnl`	Local convergence of alternating low-rank optimization methods with overrelaxation. Numer. Linear Algebra Appl. Ivan V. Oseledets, Maxim V. Rakhuba, André Uschmajew
2023	J	`jnl`	Tensor rank bounds and explicit QTT representations for the inverses of circulant matrices. Numer. Linear Algebra Appl. Lev I. Vysotsky, Maxim V. Rakhuba
2022	J	`jnl`	Automatic Differentiation for Riemannian Optimization on Low-Rank Matrix and Tensor-Train Manifolds. SIAM J. Sci. Comput. Alexander Novikov, Maxim V. Rakhuba, Ivan V. Oseledets
2022	J	`jnl`	Quantized Tensor FEM for Multiscale Problems: Diffusion Problems in Two and Three Dimensions. Multiscale Model. Simul. Vladimir A. Kazeev, Ivan V. Oseledets, Maxim V. Rakhuba, Christoph Schwab
2022	J	`jnl`	Tensor rank bounds and explicit QTT representations for the inverses of circulant matrices. CoRR Lev I. Vysotsky, Maxim V. Rakhuba
2022	A*	`conf`	Towards Practical Control of Singular Values of Convolutional Layers. NeurIPS Alexandra Senderovich, Ekaterina Bulatova, Anton Obukhov, Maxim V. Rakhuba
2022	J	`jnl`	Towards Practical Control of Singular Values of Convolutional Layers. CoRR Alexandra Senderovich, Ekaterina Bulatova, Anton Obukhov, Maxim V. Rakhuba
2021	J	`jnl`	Automatic differentiation for Riemannian optimization on low-rank matrix and tensor-train manifolds. CoRR Alexander Novikov, Maxim V. Rakhuba, Ivan V. Oseledets
2021	A*	`conf`	Cherry-Picking Gradients: Learning Low-Rank Embeddings of Visual Data via Differentiable Cross-Approximation. ICCV Mikhail Usvyatsov, Anastasia Makarova, Rafael Ballester-Ripoll, Maxim V. Rakhuba, Andreas Krause, Konrad Schindler
2021	J	`jnl`	Cherry-Picking Gradients: Learning Low-Rank Embeddings of Visual Data via Differentiable Cross-Approximation. CoRR Mikhail Usvyatsov, Anastasia Makarova, Rafael Ballester-Ripoll, Maxim V. Rakhuba, Andreas Krause, Konrad Schindler
2021	J	`jnl`	Local convergence of alternating low-rank optimization methods with overrelaxation. CoRR Ivan V. Oseledets, Maxim V. Rakhuba, André Uschmajew
2021	J	`jnl`	Robust Alternating Direction Implicit Solver in Quantized Tensor Formats for a Three-Dimensional Elliptic PDE. SIAM J. Sci. Comput. Maxim V. Rakhuba
2021	A	`conf`	Spectral Tensor Train Parameterization of Deep Learning Layers. AISTATS Anton Obukhov, Maxim V. Rakhuba, Alexander Liniger, Zhiwu Huang, Stamatios Georgoulis, Dengxin Dai, Luc Van Gool
2021	J	`jnl`	Spectral Tensor Train Parameterization of Deep Learning Layers. CoRR Anton Obukhov, Maxim V. Rakhuba, Alexander Liniger, Zhiwu Huang, Stamatios Georgoulis, Dengxin Dai, Luc Van Gool
2020	J	`jnl`	Low rank tensor approximation of singularly perturbed partial differential equations in one dimension. CoRR Carlo Marcati, Maxim V. Rakhuba, Johan E. M. Ulander
2020	A*	`conf`	T-Basis: a Compact Representation for Neural Networks. ICML Anton Obukhov, Maxim V. Rakhuba, Stamatios Georgoulis, Menelaos Kanakis, Dengxin Dai, Luc Van Gool
2020	J	`jnl`	T-Basis: a Compact Representation for Neural Networks. CoRR Anton Obukhov, Maxim V. Rakhuba, Stamatios Georgoulis, Menelaos Kanakis, Dengxin Dai, Luc Van Gool
2019	J	`jnl`	Low-rank Riemannian eigensolver for high-dimensional Hamiltonians. J. Comput. Phys. Maxim V. Rakhuba, Alexander Novikov, Ivan V. Oseledets
2019	J	`jnl`	Tensor Rank bounds for Point Singularities in R CoRR Carlo Marcati, Maxim V. Rakhuba, Christoph Schwab
2018	J	`jnl`	Alternating Least Squares as Moving Subspace Correction. SIAM J. Numer. Anal. Ivan V. Oseledets, Maxim V. Rakhuba, André Uschmajew