Parallel Distributed Processing: Explorations in the Micro-structure of Cognition

By David E. Rumelhart, J. McClelland. 2000

2,701 citations recorded by the source corpus.

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

Efficient Estimation of Word Representations in Vector Space (2013, background): “Probably the most successful concept is to use distributed representations of words [10].”
Conditional Random Fields as Recurrent Neural Networks (2015, methodology): “these parameters in a single iteration was described in Section 4, they can be learnt in the RNN setting using the standard back-propagation through time algorithm [48, 40].”
Neural Programmer-Interpreters (2015, background): “For example, the idea of using dynamically programmable networks in which the activations of one network become the weights (the program) of a second network was mentioned in the Sigma-Pi units section of the inﬂuential PDP paper (Rumelhart et al., 1986).”
Synthesizing the preferred inputs for neurons in neural networks via deep generator networks (2016, background): “Recently, advances in network architectures and training methods enabled the generation of high-dimensional realistic images [19–21 , 11 ].”
Deep learning in neural networks: An overview (2014, neutral): “The Deep Belief Network (DBN) is a stack of Restricted Boltzmann Machines (RBMs) (Smolensky, 1986), which in turn are Boltzmann Machines (BMs) (Hinton & Sejnowski, 1986) with a single layer of feature-detecting units; compare also Higher-Order BMs (Memisevic & Hinton, 2010).”
The Symbol Grounding Problem (1990, neutral): “An early rival to the symbolic model of mind appeared [36], was overcome by symbolic AI [27] and has recently re-appeared in a stronger form that is currently vying with AI to be the general theory of cognition and behavior [23] [39].”
PointCNN: Convolution On X-Transformed Points (2018, background): “, pK), with a multilayer perceptron [39], i.”
Systematic evaluation of CNN advances on the ImageNet (2016, background): “We have also tried to train sigmoid [32] network, but the initial loss never decreased.”
Forecasting stock market indexes using principle component analysis and stochastic time effective neural networks (2015, methodology): “The backpropagation algorithm has emerged as one of the most widely used learning procedures for multilayer networks [27] [28] [29].”
Deep Learning and Its Applications in Biomedicine (2018, methodology): “Subsequently, DanQ, a novel hybrid framework that combines CNN and bi-directional long short-term memory (BLSTM) RNNs, was presented to predict non-coding function de novo from sequences alone [163].”
On deep learning-based channel decoding (2017, methodology): “But it took about 45 years until the backpropagation algorithm [3] made useful applications such as handwritten ZIP code recognition possible [4].”
A Survey of Multi-View Representation Learning (2016, neutral): “Canonical Correlation Analysis is a linear multi-view representation learning algorithm, but for many scenarios of real-world multi-view data revealing nonlinearities, it is impossible for a linear embedding to capture all the properties of the multi-view data [28].”
Control of synaptic plasticity in deep cortical networks (2018, positive): “Introducing attention signals into the learning rule gives: Δwi,j = β ∙ fi(ai) ∙ fj(aj) ∙ RPE ∙ FBj (3), where FBj is the feedback from higher brain regions that gate the plasticity of synapses onto neuron j.”