Nuts and bolts for the social sciences

By Jon Elster. 1989

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Social Sciences & Humanities

Citations in the corpus

Hypercolumns for object segmentation and fine-grained localization (2014, methodology): “Malik and Perona [32] used the output of a bank of filters as a representation for texture discrimination.”
Describing Textures in the Wild (2013, methodology): “En-coding can be as simple as averaging (sum-pooling) descriptors [21], although this is often preceded by a high-Kernel dimensional sparse coding step.”
Deep learning in neural networks: An overview (2014, neutral): “The activation function f of Rectified Linear Units (ReLUs) is f(x) = x for x > 0, f(x) = 0 otherwise—compare the old concept of half-wave rectified units (Malik and Perona, 1990).”
Integral Channel Features (2009, methodology): “Panel c shows I convolved with 4 oriented Gabor filters [17] ; each channel contains edge information in I at a different orientation.”
Discriminant Saliency for Visual Recognition from Cluttered Scenes (2004, methodology): “In particular, all the results reported in this paper were achieved with the following two step procedure, based on the Malik-Perona model of texture perception [13].”
Texture analysis and segmentation using modulation features, generative models and weighted curve evolution (background): “In conjunction with both boundary-based [31], [32], [37] and region-based [30], [31], [44], [50], [57] approaches, the high dimensionality of filterbank features can lead to poor segmentations and requires dimensionality reduction, which is a problem in itself.”
A Galaxy of Texture Features (2008, background): “(18) Laplacian pyramid (− − −) Decomposing an image so that redundant information is minimised and characteristic features are thus preserved and highlighted is a common way of analysing textures.”
Residual Networks Behave Like Ensembles of Relatively Shallow Networks (2016, methodology): “This organization has been widely adopted in the computer vision and machine learning literature, from early neural networks such as the Neocognitron [4] and the traditional hand-crafted feature pipeline of Malik and Perona [15] to convolutional neural networks [13] [14].”
Scattering Representations for Recognition (2013, methodology): “4 Wavelet based texture analysis Wavelet filter banks have been used in both texture synthesis and discrimination [HB95; MP90; LM01].”
3-D Depth Reconstruction from a Single Still Image (2007, background): “Methods such as shape from shading [12; 13] and shape from texture [14; 15; 16] generally assume uniform color and/or texture, 1 and hence would perform very poorly on the complex, unconstrained, highly textured images that we consider.”
Two Distinct Mechanisms of Suppression in Human Vision (2005, methodology): “An alternative application for such a narrowly tuned long-range suppression is texture segmentation, as in the models by Malik and Perona (1990) and by Li (2000).”
Beyond Bounding Boxes: Precise Localization of Objects in Images (2015, methodology): “Malik and Perona [46] used the output of a bank of filters as a representation for texture discrimination.”
The Art of Good Hope (2004, background): “…attitude akin to, though perhaps also partly composed of, beliefs and desires (Bovens 1999; Pettit this volume), an emotion (Drahos this volume; Elster 1989), a disposition or capacity (Gravlee 2000), a process or activity (Snyder 1995; V. Braithwaite this volume), or finally, some combination…”