Serial Order: A Parallel Distributed Processing Approach

By Michael I. Jordan. 1997

Subfields: Neuroscience, Cognitive Science & Psychology

How researchers see this paper

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

In Neuroscience

In Cognitive Science & Psychology

Citations in the corpus

TensorFlow: A system for large-scale machine learning (2016, neutral): “This pattern works for training simple feed-forward neural networks, but fails for more advanced models, such as recurrent neural networks, which contain loops [39]; adversarial networks, in which two related networks are trained alternately [26]; and reinforcement learning models, where the loss fu”
Deep learning in neural networks: An overview (2014, neutral): “In fact, some popular RNN algorithms restricted credit assignment to a single step backwards (Elman, 1990; Jordan, 1986, 1997), also inmore recent studies (Jaeger, 2001, 2004; Maass et al., 2002).”
The History Began from AlexNet: A Comprehensive Survey on Deep Learning Approaches (2018, methodology): “On the other hand, the outputs from output unit are used as inputs with the inputs of hidden layer in Jordan network [130].”
Doing without schema hierarchies: a recurrent connectionist approach to normal and impaired routine sequential action. (2004, background): “Beginning with the pioneering work of Jordan (1986b) and Elman (1990, 1991, 1993), numerous studies have demonstrated the ability of such networks to produce sequential behavior resembling that of humans in a variety of domains, including spoken word comprehension and production (Christiansen,…”
Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network (2018, methodology): “sentially what is dubbed “deep networks” and “deep learning” today. 4 foundation of Back Propagation and used to train the weights and time delays of these networks to perform a wide variety of tasks [11, 25, 45, 46, 47, 48]. We will rely on Back Propagation Through Time for training the systems ana”
Neural networks with a self-refreshing memory: Knowledge transfer in sequential learning tasks without catastrophic forgetting (2000, unknown): “In further research, the self-refreshing memory model will be extended to pseudorehearsal of ordered time-series which are stored by recurrent networks (e.g. Jordan 1986, Elman 1990, Reiss and Taylor 1991, Ans et al .”
Deep Learning in Protein Structural Modeling and Design (2020, methodology): “Recurrent Neural Networks Recurrent architectures are based on applying several iterations of the same function along a sequential input.(52) This can be seen as an unfolded architecture, and it has been widely used to process sequential data, such as time series data and written text”
Wave physics as an analog recurrent neural network (2019, methodology): “[15] Michael I Jordan, “Serial order: A parallel distributed processing approach,” in Advances in psychology, Vol.”
Deep Learning for Time Series Forecasting: Tutorial and Literature Survey (2020, background): “The core idea consists of connecting recurrently the NNs’ hidden units back to themselves with a time delay [94, 95].”
Meta-learning of Sequential Strategies (2019, methodology): “If we use a memory-based architecture, such as a recurrent neural network [Elman, 1990, Jordan, 1997, Hochreiter et al., 2001, Kolen and Kremer, 2001, Graves, 2012], then the function f conforms to the interface where π t ∈ ∆( X ) is the current prediction vector, m t and m t − 1 ∈ M are the…”
Neural Program Lattices (2016, methodology): “This work has focused largely on more standard RNN architectures, starting with Cottrell & Tsung (1993), which showed that the standard RNN architectures at the time (Jordan, 1997; Elman, 1990) could successfully generalize to test samples approximately 5 times as long as those seen during training,”
Becoming syntactic. (2006, background): “This feeding back of produced output is particularly important for production because it helps the model keep track of the kind of sentence that it is producing and where it is in the sentence (Jordan, 1986).”
A Survey of Deep Active Learning (2020, background): “86, [129] introduced backpropagation into the optimization of neural networks, which laid the foundation for the subsequent rapid development of DL. In the same year, Recurrent Neural Networks (RNNs) [75] were first proposed. In 1998, the LeNet [92] network made its first appearance, representing on”
TAR-cointegration neural network model: An empirical analysis of exchange rates and stock returns (2010, methodology): “…NN model (Elman, 1990), the hidden neuron outputs at the previous time step is fed back to its inputs through time lag units; whereas, in Jordan RNN model (Jordan, 1986), the output units are connected to inputs with time delay so that the network output in time t 1 is also information on time t.”
Extracting comprehensible models from trained neural networks (1996, methodology): “Hanson (1990) used a hierarchical-clustering method to identify clusters, while others have used principal-components analysis (Jolliffe, 1986) (and the related method of canonical-discriminant analysis) for this task (Dennis & Phillips, 1991; Elman, 1991).”
Machine Learning for Sequential Data: A Review (2002, unknown): “by Jordan [11].”