30 years of adaptive neural networks: perceptron, Madaline, and backpropagation

By Bernard Widrow, Michael A. Lehr. 1990

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

Born to Learn: the Inspiration, Progress, and Future of Evolved Plastic Artificial Neural Networks (2017, background): “Because networks perform correctly only with speciﬁc weight conﬁgurations, learning rules are an integral part of most neural networks; see Widrow and Lehr (1990) for an early review.”
Automatic differentiation in machine learning: a survey (2015, background): “Examples include the training of artificial neural networks (Widrow and Lehr, 1990), conditional random fields (Vishwanathan et al, 2006), natural language processing (Finkel et al, 2008), and computer vision (Parker, 2010).”
Automatic Differentiation Variational Inference (2016, methodology): “For example, the backpropagation algorithm, rediscovered independently many times, is a form of automatic differentiation for neural network weights (Widrow and Lehr, 1990).”
A review of automatic differentiation and its efficient implementation (2018, background): “In machine learning, gradients play a crucial role in the training of neural networks (Widrow & Lehr, 1990).”
On-line learning algorithms for locally recurrent neural networks (1999, methodology): “This notation, which is a generalization of that used in [48] for static MLP and in [8] and [11] for FIRMLP, is appropriate in this case where complex architectures of different kinds are defined and compared.”
Univariate modelling of monthly maximum temperature time series over northeast India: neural network versus Yule–Walker equation based approach (2011, background): “Further details of MLP can be found in such surveys as Sarkar (1995) and Widrow and Lehr (1990).”
NeuroAnimator: fast neural network emulation and control of physics-based models (1998, background): “Research in neural networks dates as far back as the 1 9 6 0 ~ ~ when Widrow and co-workers proposed networks they called adalines (Widrow and Lehr, 1990).”
Modeling, Inference and Optimization With Composable Differentiable Procedures (2016, background): “Reverse mode differentiation has been independently discovered many times in various quantitative fields [9], and even several times within machine learning [114].”
Neural Modularity Helps Organisms Evolve to Learn New Skills without Forgetting Old Skills (2015, unknown): “Some learning algorithms, such as backpropagation [6, 7], require a correct output (e.”
Pattern Recognition and Machine Learning (2006, unknown): “The functional form of the model was the same as for the perceptron, but a different approach to training was adopted (Widrow and Hoff, 1960; Widrow and Lehr, 1990).”
Evolutionary and Computational Advantages of Neuromodulated Plasticity (2008, background): “In the field of feed-forward neural networks for classification tasks and in many early approaches to learning in neural networks, supervised learning and forms of gradient descent – which are not however a focus in this thesis – have been proposed (Widrob and Lehr, 1990).”