Zobrazeno 1 - 10
of 24 847
pro vyhledávání: '"As, Gorbunov"'
Autor:
Demidovich, Yury, Ostroukhov, Petr, Malinovsky, Grigory, Horváth, Samuel, Takáč, Martin, Richtárik, Peter, Gorbunov, Eduard
Non-convex Machine Learning problems typically do not adhere to the standard smoothness assumption. Based on empirical findings, Zhang et al. (2020b) proposed a more realistic generalized $(L_0, L_1)$-smoothness assumption, though it remains largely
Externí odkaz:
http://arxiv.org/abs/2412.02781
Autor:
Ponkshe, Kaustubh, Singhal, Raghav, Gorbunov, Eduard, Tumanov, Alexey, Horvath, Samuel, Vepakomma, Praneeth
Low-rank adapters have become a standard approach for efficiently fine-tuning large language models (LLMs), but they often fall short of achieving the performance of full fine-tuning. We propose a method, LoRA Silver Bullet or LoRA-SB, that approxima
Externí odkaz:
http://arxiv.org/abs/2411.19557
Autor:
Reinold, Anneke, Berger, Lucas, Raczkowski, Marcin, Zhao, Zhiying, Kohama, Yoshimitsu, Gen, Masaki, Gorbunov, Denis I., Skourski, Yurii, Zherlitsyn, Sergei, Assaad, Fakher F., Lorenz, Thomas, Wang, Zhe
We report on a comprehensive thermodynamic study of a quasi-two-dimensional (quasi-2D) quantum magnet Cu$_2$(OH)$_3$Br which in the 2D layer can be viewed as strongly coupled alternating antiferromagnetic and ferromagnetic chains. In an applied magne
Externí odkaz:
http://arxiv.org/abs/2411.09541
We study domain walls (DWs) arising in field theories where $Z_2$-symmetry is spontaneously broken by a scalar expectation value decreasing proportionally to the Universe temperature. The energy density of such melting DWs redshifts sufficiently fast
Externí odkaz:
http://arxiv.org/abs/2410.21971
We provide the first proof of convergence for normalized error feedback algorithms across a wide range of machine learning problems. Despite their popularity and efficiency in training deep neural networks, traditional analyses of error feedback algo
Externí odkaz:
http://arxiv.org/abs/2410.16871
Autor:
Gorbunov, Evgeny A., Bacchini, Fabio, Zhdankin, Vladimir, Werner, Gregory R., Begelman, Mitchell C., Uzdensky, Dmitri A.
We present the largest 3D Particle-in-Cell shearing-box simulations of turbulence driven by the magnetorotational instability, for the first time employing the realistic proton-to-electron mass ratio. We investigate the energy partition between relat
Externí odkaz:
http://arxiv.org/abs/2410.02872
Autor:
Heinze, L., Kotte, T., Rausch, R., Demuer, A., Luther, S., Feyerherm, R., Ammerlaan, A. A. L. N., Zeitler, U., Gorbunov, D. I., Uhlarz, M., Rule, K. C., Wolter, A. U. B., Kühne, H., Wosnitza, J., Karrasch, C., Süllow, S.
We report an extensive high-field study of atacamite Cu$_2$Cl(OH)$_3$, a material realization of quantum sawtooth chains with weak interchain couplings, in continuous and pulsed magnetic fields up to 58 T. In particular, we have mapped the entropy la
Externí odkaz:
http://arxiv.org/abs/2410.01947
In this study, we probe the coupling between magnetism and lattice dynamics in UO2, a 3k antiferromagnet that undergoes magnetic ordering below its Neel temperature (TN) of 30.8 K. Ultrasound measurements provide insights into the interplay between t
Externí odkaz:
http://arxiv.org/abs/2409.20228
Autor:
Gorbunov, Eduard, Tupitsa, Nazarii, Choudhury, Sayantan, Aliev, Alen, Richtárik, Peter, Horváth, Samuel, Takáč, Martin
Due to the non-smoothness of optimization problems in Machine Learning, generalized smoothness assumptions have been gaining a lot of attention in recent years. One of the most popular assumptions of this type is $(L_0,L_1)$-smoothness (Zhang et al.,
Externí odkaz:
http://arxiv.org/abs/2409.14989
Controlling the spectral norm of the Jacobian matrix, which is related to the convolution operation, has been shown to improve generalization, training stability and robustness in CNNs. Existing methods for computing the norm either tend to overestim
Externí odkaz:
http://arxiv.org/abs/2409.11859