Zobrazeno 1 - 10
of 150
pro vyhledávání: '"Campbell, Ryan P."'
Autor:
Campbell, Ryan, Wadsworth, Jennifer
A recent development in extreme value modeling uses the geometry of the dataset to perform inference on the multivariate tail. A key quantity in this inference is the gauge function, whose values define this geometry. Methodology proposed to date for
Externí odkaz:
http://arxiv.org/abs/2412.05195
Autor:
Campbell, Ryan, Lojo, Nelson, Viswanadha, Kesava, Tryggestad, Christoffer Grondal, Sun, Derrick Han, Vijapurapu, Sriteja, Rolfsen, August, Sahai, Anant
In-Context Learning (ICL) is a phenomenon where task learning occurs through a prompt sequence without the necessity of parameter updates. ICL in Multi-Headed Attention (MHA) with absolute positional embedding has been the focus of more study than ot
Externí odkaz:
http://arxiv.org/abs/2411.03945
Autor:
Campbell, Ryan, Yoon, Junsang
This paper investigates the impact of using gradient norm reward signals in the context of Automatic Curriculum Learning (ACL) for deep reinforcement learning (DRL). We introduce a framework where the teacher model, utilizing the gradient norm inform
Externí odkaz:
http://arxiv.org/abs/2312.13565
In-context learning (ICL) has revolutionized the capabilities of transformer models in NLP. In our project, we extend the understanding of the mechanisms underpinning ICL by exploring whether transformers can learn from sequential, non-textual functi
Externí odkaz:
http://arxiv.org/abs/2312.12655
We use a functional analogue of the quantile function for probability measures admitting a continuous Lebesgue density on $\mathbb{R}^d$ to characterise the class of non-trivial limit distributions of radially recentered and rescaled multivariate exc
Externí odkaz:
http://arxiv.org/abs/2310.06130
Autor:
Campbell, Ryan J., Keys, P H., Mathioudakis, M., Woeger, F., Schad, T. A., Tritschler, A., de Wijn, A. G., Smitha, H. N., Beck, C. A., Christian, D J., Jess, D. B., Erdelyi, R.
We present the first quiet Sun spectropolarimetric observations obtained with the Visible SpectroPolarimeter (ViSP) at the $4-$m Daniel K. Inouye Solar Telescope (DKIST). We recorded observations in a wavelength range that includes the magnetically s
Externí odkaz:
http://arxiv.org/abs/2309.05836
Autor:
Campbell, Ryan James, Gafeira, Ricardo, Mathioudakis, Mihalis, Noda, Carlos Quintero, Collados, Manuel
Synthetic observations produced from radiative magnetohydronamic simulations have predicted that higher polarization fractions in the quiet solar photosphere would be revealed by increasing the total integration time of observations at GREGOR resolut
Externí odkaz:
http://arxiv.org/abs/2301.05591
Autor:
Wadsworth, Jennifer, Campbell, Ryan
A geometric representation for multivariate extremes, based on the shapes of scaled sample clouds in light-tailed margins and their so-called limit sets, has recently been shown to connect several existing extremal dependence concepts. However, these
Externí odkaz:
http://arxiv.org/abs/2208.14951
Autor:
Campbell, Ryan J., Mathioudakis, Mihalis, Collados, Manuel, Keys, Peter H., Ramos, Andrés Asensio, Nelson, Chris J., Kuridze, David, Reid, Aaron
Publikováno v:
A&A 647, A182 (2021)
While the longitudinal field that dominates photospheric network regions has been studied extensively, small scale transverse fields have recently been found to be ubiquitous in the quiet internetwork photosphere. Few observations have captured how t
Externí odkaz:
http://arxiv.org/abs/2102.00942
Machine learning models are vulnerable to adversarial attacks. One approach to addressing this vulnerability is certification, which focuses on models that are guaranteed to be robust for a given perturbation size. A drawback of recent certified mode
Externí odkaz:
http://arxiv.org/abs/2010.02508