Neural Networks for Pattern Recognition

By Franziska Wulf. 1993

How researchers see this paper

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

In Model Architectures

Citations in the corpus

Deep Residual Learning for Image Recognition (2015, background): “Shortcut connections [2, 34, 49] are those skipping one or more layers.”
Generating Sequences With Recurrent Neural Networks (2013, methodology): “The idea of mixture density networks [2, 3] is to use the outputs of a neural network to parameterise a mixture distribution.”
Elucidating the Design Space of Diffusion-Based Generative Models (2022, background): “For example, it is advisable to keep input and output signal magnitudes fixed to, e.g., unit variance, and to avoid large variation in gradient magnitudes on a per-sample basis [5, 21].”
Understanding deep image representations by inverting them (2014, background): “Others [1] proposed to learn a second neural network to act as the inverse of the original one, but this is complicated by the fact that the inverse is usually not unique.”
Inverting Visual Representations with Convolutional Networks (2015, methodology): “These include works making use of backpropagation or sampling [15, 16, 18, 27, 9, 25] and, most similar to our approach, other neural networks [2].”
Compressing Neural Networks with the Hashing Trick (2015, background): “Let L denote the loss function for training the neural network, e.g. cross entropy or the quadratic loss (Bishop, 1995).”
Gaussian Ranking by Matrix Factorization (2015, neutral): “σ(x) = 1 1 + exp(−x) (12) with appropriate scaling: probit(s) ≈ σ(s √ 8/π), see eg [1].”
Methods for interpreting and understanding deep neural networks (2017, methodology): “The technique is easy to implement for a deep neural network, since the gradient can be computed using backpropagation [11, 41].”
Explaining nonlinear classification decisions with deep Taylor decomposition (2015, background): “ral networks [5, 6, 7], boosting [8], or random forests [9].”
On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (2015, methodology): “Neural networks are trained by backpropagation using stochastic gradient descent with mini batches of size 25 and using a softmax objective [6].”
Learning Robust Representations by Projecting Superficial Statistics Out (2018, methodology): “We write F TG F G + λI back to the previous equation, With some linear algebra and Kailath Variant (Bishop et al., 1995), we can have: is a result of a heteroscadestic regression method where λ can be estimated through maximum likelihood estimation (MLE) (Wang et al., 2017), which completes the…”
Detecting and Correcting for Label Shift with Black Box Predictors (2018, background): “When q ( y ) is known, label shift simpliﬁes to the problem of changing base rates (Bishop, 1995; Elkan, 2001).”
Interval Coded Scoring: a toolbox for interpretable scoring systems (2018, methodology): “Well-known approaches include Support Vector Machines (SVM) (Vapnik, 1995), Naive Bayes (NB) classiﬁers (Lewis, 1998), decision trees (DT) (Quinlan, 1986), random forests (RF) (Ho, 1995) and neural networks (NN) (Bishop, 1995).”
Compressing Convolutional Neural Networks in the Frequency Domain (2015, background): “During training, the hashed weights can be learned with simple back-propagation [3]—the gradient of a hash bucket value is the sum of gradients of all hashed frequency components in that bucket.”
How to Explain Individual Classification Decisions (2009, methodology): “1 In the neural network literature also solely an ensemble view was taken in algorithms like input pruning (e.g. Bishop, 1995; LeCun, Bottou, Orr, and M¨uller, 1998).”
On the mathematical foundations of learning (2001, neutral): “[6] and page 41 in [28]).”