Zobrazeno 1 - 10
of 20
pro vyhledávání: '"Shumaylov, Zakhar"'
Autor:
Canizares, Priscilla, Murari, Davide, Schönlieb, Carola-Bibiane, Sherry, Ferdia, Shumaylov, Zakhar
Hamilton's equations of motion form a fundamental framework in various branches of physics, including astronomy, quantum mechanics, particle physics, and climate science. Classical numerical solvers are typically employed to compute the time evolutio
Externí odkaz:
http://arxiv.org/abs/2410.18262
Autor:
Shumaylov, Zakhar, Zaika, Peter, Rowbottom, James, Sherry, Ferdia, Weber, Melanie, Schönlieb, Carola-Bibiane
The quest for robust and generalizable machine learning models has driven recent interest in exploiting symmetries through equivariant neural networks. In the context of PDE solvers, recent works have shown that Lie point symmetries can be a useful i
Externí odkaz:
http://arxiv.org/abs/2410.02698
Data-driven Riemannian geometry has emerged as a powerful tool for interpretable representation learning, offering improved efficiency in downstream tasks. Moving forward, it is crucial to balance cheap manifold mappings with efficient training algor
Externí odkaz:
http://arxiv.org/abs/2410.01950
Variational regularisation is the primary method for solving inverse problems, and recently there has been considerable work leveraging deeply learned regularisation for enhanced performance. However, few results exist addressing the convergence of s
Externí odkaz:
http://arxiv.org/abs/2402.01052
An emerging new paradigm for solving inverse problems is via the use of deep learning to learn a regularizer from data. This leads to high-quality results, but often at the cost of provable guarantees. In this work, we show how well-posedness and con
Externí odkaz:
http://arxiv.org/abs/2310.05812
Autor:
Shumailov, Ilia, Shumaylov, Zakhar, Zhao, Yiren, Gal, Yarin, Papernot, Nicolas, Anderson, Ross
Stable Diffusion revolutionised image creation from descriptive text. GPT-2, GPT-3(.5) and GPT-4 demonstrated astonishing performance across a variety of language tasks. ChatGPT introduced such language models to the general public. It is now clear t
Externí odkaz:
http://arxiv.org/abs/2305.17493
Autor:
Letey, Mary I., Shumaylov, Zakhar, Agocs, Fruzsina J., Handley, Will J., Hobson, Michael P., Lasenby, Anthony N.
Publikováno v:
Phys. Rev. D 109, 123502 (2024)
We discuss the challenges of motivating, constructing, and quantizing a canonically normalized inflationary perturbation in spatially curved universes. We show that this has historically proved challenging due to the interaction of nonadiabaticity wi
Externí odkaz:
http://arxiv.org/abs/2211.17248
Autor:
Shumaylov, Zakhar, Handley, Will
Publikováno v:
Phys. Rev. D 105, 123532 (2022)
We investigate the primordial power spectra for general kinetic inflation models that support a period of kinetic dominance in the case of curved universes. We present derivations of the Mukhanov-Sasaki equations with a nonstandard scalar kinetic Lag
Externí odkaz:
http://arxiv.org/abs/2112.07547
Autor:
Shumailov, Ilia, Shumaylov, Zakhar, Kazhdan, Dmitry, Zhao, Yiren, Papernot, Nicolas, Erdogdu, Murat A., Anderson, Ross
Machine learning is vulnerable to a wide variety of attacks. It is now well understood that by changing the underlying data distribution, an adversary can poison the model trained with it or introduce backdoors. In this paper we present a novel class
Externí odkaz:
http://arxiv.org/abs/2104.09667
Autor:
Mukherjee, Subhadip, Dittmer, Sören, Shumaylov, Zakhar, Lunz, Sebastian, Öktem, Ozan, Schönlieb, Carola-Bibiane
We consider the variational reconstruction framework for inverse problems and propose to learn a data-adaptive input-convex neural network (ICNN) as the regularization functional. The ICNN-based convex regularizer is trained adversarially to discern
Externí odkaz:
http://arxiv.org/abs/2008.02839