Zobrazeno 1 - 10
of 130
pro vyhledávání: '"Jonasson, Johan"'
A first step is taken towards understanding often observed non-robustness phenomena of deep neural net (DNN) classifiers. This is done from the perspective of Boolean functions by asking if certain sequences of Boolean functions represented by common
Externí odkaz:
http://arxiv.org/abs/2308.09374
Subsampling is commonly used to overcome computational and economical bottlenecks in the analysis of finite populations and massive datasets. Existing methods are often limited in scope and use optimality criteria (e.g., A-optimality) with well-known
Externí odkaz:
http://arxiv.org/abs/2304.03019
Autor:
Bergstrand, Björn, Jonasson, Johan
Syftet med studien är att undersöka vilka preventiva och behandlande effektiva insatser det finns för att motverka mobbning i skolmiljö. Redan under tidigt 70-tal började mobbning betraktas som ett socialt problem. Än idag är mobbning i svensk
Externí odkaz:
http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-201873
Publikováno v:
Journal of Machine Learning Research 24 (2023) 1-53
The elastic net combines lasso and ridge regression to fuse the sparsity property of lasso with the grouping property of ridge regression. The connections between ridge regression and gradient descent and between lasso and forward stagewise regressio
Externí odkaz:
http://arxiv.org/abs/2202.02146
Autor:
Jonasson, Johan, Elmgren, Magnus
The main objective of this study is to explore how preschool teachers describe their experiences of working with children in need of special support. The empirical data used in this qualitative study was reached by conducting phone interviews with se
Externí odkaz:
http://urn.kb.se/resolve?urn=urn:nbn:se:mau:diva-31271
The presence of mislabeled observations in data is a notoriously challenging problem in statistics and machine learning, associated with poor generalization properties for both traditional classifiers and, perhaps even more so, flexible classifiers l
Externí odkaz:
http://arxiv.org/abs/2112.08102
Autor:
Jonasson, Johan, Magnusson, Måns
We consider nearest neighbor weighted random walks on the $d$-dimensional box $[n]^d$ that are governed by some function $g:[0,1] \ra [0,\iy)$, by which we mean that standing at $x$, a neighbor $y$ of $x$ is picked at random and the walk then moves t
Externí odkaz:
http://arxiv.org/abs/2101.10004
Publikováno v:
Electron. Commun. Probab. 22 (2017), Paper No. 21
In this paper we study the existence phase transition of the random fractal ball model and the random fractal box model. We show that both of these are in the empty phase at the critical point of this phase transition.
Externí odkaz:
http://arxiv.org/abs/2004.00393
Publikováno v:
Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 108:341-351, 2020
Recently, new methods for model assessment, based on subsampling and posterior approximations, have been proposed for scaling leave-one-out cross-validation (LOO) to large datasets. Although these methods work well for estimating predictive performan
Externí odkaz:
http://arxiv.org/abs/2001.00980
Publikováno v:
Thirty-sixth International Conference on Machine Learning, PMLR 97:4244-4253, 2019
Model inference, such as model comparison, model checking, and model selection, is an important part of model development. Leave-one-out cross-validation (LOO) is a general approach for assessing the generalizability of a model, but unfortunately, LO
Externí odkaz:
http://arxiv.org/abs/1904.10679