Scale-Equivariant Steerable NetworksIvan SosnovikMicha SzmajaArnold SmeuldersInternational Conference on Learning Representations
Many robust-regression estimators are regression and scale equivariant.Normality tests in linear regressions using these estimators control size exactly.In a simulation study, we show that tests based on least squares can be biased.关键词: Normality test Linear regression Regression and scale equivariant...
(1989). Minimum risk equivariant estimator: Estimating the mean of an inverse Gaussian distribution with known coefficient of variation, Communications in Statistics-Theory and Methods , 18 , 189–197. MATH MathSciNetHirano, K. and Iwase, K. (1989b). Minimum risk scale equivariant estimator: ...
Nerfstudio Implementation of RENI++: A Rotation-Equivariant, Scale-Invariant, Natural Illumination Prior - JADGardner/ns_reni
Lehmann (Theory of Point Estimation, 1983, Wiley, New York) discusses the problem of marginal equivariant estimation of the parameters of a location-scale model. In this paper, we develop the procedure of simultaneous equivariant estimation of the parameters and extend the results discussed in Lehma...
In this work, we present Allegro, an equivariant deep-learning approach that retains the high accuracy of the recently proposed class of equivariant MPNNs15,26,27,37,39,40while combining it with strict locality and thus the ability to scale to large systems. We demonstrate that Allegro not onl...
In this work, we propose modeling the proximal operators of unrolled neural networks with scale-equivariant convolutional neural networks in order to improve the data-efficiency and robustness to drifts in scale of the images that might stem from the variability of patient anatomies or change in ...
However, no method has focused on the difference in scale within an image caused by a 3-dimensional to 2-dimensional projection. We proposed a new scale-equivariant convolution method that focuses on the relationship between the object distance and scale ratio in the image.Marumo, Hidetaka...
In both classification and segmentation tasks, the scale-equivariant architectures improve dramatically the generalization to unseen scales compared to a convolutional baseline. Besides, in our experiments morphological scale-spaces outperformed the Gaussian scale-space in geometrical tasks....
The scale-equivariant convolutional neural network includes a convolutional layer. The convolutional layer is configured to provide a convolution output based on a plurality of steerable filters of the convolutional layer and a convolution input. The convolution input is based on the input image and ...