Neural networks for pattern recognition

By Charles M. Bishop. 1995

How researchers see this paper

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

Citations in the corpus

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges. (2019, background): “i=1 Ei (11) A potential shortcoming of Eq.”
Visualizing Deep Convolutional Neural Networks Using Natural Pre-images (2015, methodology): “ctivations of max-pooling layers from the direct CNN evaluation to the transposed one, therefore copying rather than inferring this geometric information during reconstruction. A related line of work [1,7] is to learn a second neural network to act as the inverse of the original one. This is difﬁcul”
PowerPlay: Training an Increasingly General Problem Solver by Continually Searching for the Simplest Still Unsolvable Problem (2011, background): “pesh Srivastava (a separate report on this is in preparation). A ﬁrst experimental setup was limited to patt ern classiﬁcation (Section 3.1.1): s is the weight 13 matrix of a ﬁxed-size feedforward NN [2] which maps 2-dimensional real-valued input vectors from the unit square [0,1)× [0,1)to binary la”
Effective Use of Word Order for Text Categorization with Convolutional Neural Networks (2014, methodology): “he output of the pooling layer z at each location by multiplying (1+jzj2) 1=2. 3.2 Baseline methods For comparison, we tested SVM with the linear kernel and fully-connected neural networks (see e.g., Bishop (1995)) with bag-of-n-gram vectors as input. To experiment with fully-connected neural nets, ”
Task-Agnostic Continual Learning Using Online Variational Bayes With Fixed-Point Updates (2020, background): “Bayesian inference (Gelman et al., 2013; Bishop, 1995) requires a joint probability distribution over the target set D and the model parameters θ given the input set X .”
Discriminative Training of a Neural Network Statistical Parser (2004, methodology): “Cross-entropy error ensures that the minimum of the error function converges to the desired probabilities as the amount of training data increases (Bishop, 1995), so the minimum for any given dataset is considered an estimate of the true probabilities.”
Inductive biases for deep learning of higher-level cognition (2020, background): “There are many ways to encode such biases—e.g. explicit regularisation objectives (Bishop et al., 1995; Bishop, 1995; Srivastava et al., 2014; Kukačka et al., 2017; Zhang et al., 2017), architectural constraints (Yu and Koltun, 2015; Long et al., 2015; Dumoulin and Visin, 2016; He et al., 2016;…”
Deep UQ: Learning deep neural network surrogate models for high dimensional uncertainty quantification (2018, methodology): “In this work, we propose a systematic approach for constructing surrogate models using deep neural networks (DNNs) [37, 38, 39, 40].”
Propensity score estimation: machine learning and classification methods as alternatives to logistic regression (2010, background): “First, they are designed to deal with highdimensional data, each value of which may have only a slight effect on the probability of class membership, but which, as a group, can classify accurately [15].”
Nonparametric Density Estimation: Toward Computational Tractability (2003, background): “In general, density estimation provides a classical basis across statistics for virtually any kind of data analysis in principle, including clustering, classification, regression, time series analysis, active learning, and so on [7, 1 ].”
A simple and fast secondary structure prediction method using hidden neural networks (2005, methodology): “To train the YASPIN NN, we used the on-line backpropagation algorithm and 6-fold cross-validation (Bishop, 1995).”
Continuous space language models (2007, background): “(6)) can be seen as a smoothed version of the winner-takes-all activation model in which the unit with the largest input has output +1 while all other units have output 0 (Bridle, 1990; Bishop, 1995).”