A logical calculus of the ideas immanent in nervous activity

By W. McCulloch, W. Pitts. 1990

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

Gaussian Error Linear Units (GELUs) (2016, methodology): “Early artificial neurons utilized binary threshold units (Hopfield, 1982; McCulloch & Pitts, 1943).”
Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning (2017, background): “Deep Learning is a new branch of machine learning that makes use of neural networks, a concept which dates back to 1943 [47],”
Deep learning in neural networks: An overview (2014, methodology): “So far we have focused on Deep Learning (DL) in supervised or unsupervised NNs.”
A practical tutorial on autoencoders for nonlinear feature fusion: Taxonomy, models, software and guidelines (2018, background): “McCulloch and Pitts cell [1] was proposed back in 1943, and the Hebb rule [2] that the Perceptron [3] is founded on was stated in 1949.”
Are Markets Truly Efficient? Experiments Using Deep Learning Algorithms for Market Movement Prediction (2018, neutral): “Neural networks (NNs), the primary methodological tool used in our analysis, have been in existence since the 1940s, beginning with the early work of [6], but have become immensely popular in the last decade due to the increase in computing power and the explosion of available data.”
Opportunities and obstacles for deep learning in biology and medicine (2017, background): “Neural networks were first proposed in 1943 [6] as a model for how our brains process information.”
Ternary Feature Masks: continual learning without any forgetting (2020, background): “A no-forgetting idea is discussed in the appendices of their manuscript with a note on binary masks, connecting the removal of plasticity to the inhibitory synapses idea [30].”
Bitwise Neural Networks (2016, methodology): “…long been known that any Boolean function, which takes binary values as input and produces binary outputs as well, can be represented as a bitwise network with one hidden layer (McCulloch & Pitts, 1943), for example, by merely memorizing all the possible mappings between input and output patterns.”
Learning values across many orders of magnitude (2016, background): “An important special case is when f θ is a neural network [McCulloch and Pitts, 1943, Rosenblatt, 1962], which are often trained with a form of SGD [Rumelhart et al., 1986], with hyperparameters that interact with the scale of the loss.”
Recurrent Relational Networks (2017, background): “There is rich literature on combining symbolic reasoning and logic with sub-symbolic distributed representations which goes all the way back to the birth of the idea of parallel distributed processing McCulloch and Pitts [1943].”
The Utility of Sparse Representations for Control in Reinforcement Learning (2018, background): “Certain activation functions—such as linear threshold units (LTU) (McCulloch and Pitts 1943) and rectiﬁed linear units (ReLU) (Glorot, Bordes, and Ben-gio 2011)—naturally provide some level of sparsity, but of course provide no such guarantees.”
Gaussian Process Kernels for Pattern Discovery and Extrapolation (2013, methodology): “The ﬁrst machine learning models, such as the perceptron (Rosenblatt, 1962), were based on a simple model of a neuron (McCulloch & Pitts, 1943).”
Deep learning for computational biology (2016, neutral): “…relatively low number of 10,700 training samples in combination with the model complexity, this method achieved substantially higher prediction accuracy of splicing activity compared to simpler approaches, and in particular was able to identify rare mutations implicated in splicing misregulation.”
Artificial Intelligence (AI) and Internet of Medical Things (IoMT) Assisted Biomedical Systems for Intelligent Healthcare (2022, unknown): “regression and has medium accuracy.”