Associative Storage and Retrieval of Digital Information in Networks of Adaptive “Neurons”

Widrow, Bernard; Hoff, Marcian E.

doi:10.1007/978-1-4684-1716-6_25

Cited by 51 publications

(34 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this adaptive process, the connection strengths between the active elements of the network are gradually modified until the network exhibits a desired behavior. A widely used adaptive method is the backpropagation of errors technique, as discussed by Rumelhart, Hinton, and Williams [19] which is a generalization of the delta rule developed by Widrow and Hoff [43], [44], and le Cun [45]. The method seeks to minimize the error in the output of the network, as compared to a target, or desired, response.…”

Section: Neural Networkmentioning

confidence: 99%

Pre-production results demonstrating multiple-system models for yield analysis

Rietman¹,

Friedman²,

Lory³

1997

IEEE Trans. Semicond. Manufact.

View full text Add to dashboard Cite

We have assembled an integrated view of the entire via manufacturing process. This integrated study includes five key plasma processes that culminate in the production of vias on CMOS wafers. There are essentially no linear cross-correlations between the processing steps and there are no linear correlations between the individual process steps and the yield for vias, as measured by the resistance between metal-one (M1) and metaltwo (M2).Using a neural network, we demonstrate that the key processing steps to determine the M1M2 resistance are the thick oxide deposition and the anisotropic via etch. Of lesser significance are the etchback planarization, an isotropic etch and plasma enhanced tetra-ethoxy siliane (PETEOS) deposition. Keeping in mind that there are five processing steps, the numerical value of M1M2 resistance can be predicted ahead of time, before completion of all five processes. This prediction can be done to an accuracy of about 1 : By using adaptive neural networks, the intelligent agents can modify their predictive behavior with respect to process changes effected by the engineering staff. Our pre-production demonstration suggests that these programs could be used in feedback and feedforward control for production yield.

show abstract

Section: Neural Networkmentioning

confidence: 99%

Pre-production results demonstrating multiple-system models for yield analysis

Rietman¹,

Friedman²,

Lory³

1997

IEEE Trans. Semicond. Manufact.

View full text Add to dashboard Cite

show abstract

“…4a). The algorithm was originally introduced in [10][11][12]. Briefly, LMS computes a prediction error for an affine decoder (i.e.…”

Section: Online Lms Algorithmmentioning

confidence: 99%

Stable task information from an unstable neural population

Loback

Rule

Raman

et al. 2019

Preprint

View full text Add to dashboard Cite

Over days and weeks, neurons in mammalian sensorimotor cortex have been found to continually change their activity patterns during performance of a learned sensorimotor task, with no detectable change in behaviour. This challenges classical theories of neural circuit function and memory, which posit that stable engrams underlie stable learned behavior [1,2]. Using existing experimental data we show that a simple linear readout can accurately recover behavioural variables, and that fixed linear weights can approximately decode behaviour over many days, despite significant changes in neural tuning. This implies that an appreciable fraction of ongoing activity reconfiguration occurs in an approximately linear subspace of population activity. We quantify the amount of additional plasticity that would be required to compensate for reconfiguration, and show that a biologically plausible local learning rule can achieve good decoding accuracy with physiologically achievable rates of synaptic plasticity.

show abstract

“…Новиков доказал сходимость предложенного метода обучения нейрона на основе правил Хэбба [20], при условии, что выборка объектов линейно разделима. Впоследствии было предложено несколько аналогичных правил как для обучением с учителем [21][22][23], так и без учителя [24][25][26][27][28].…”

Section: обучение нейронных сетейunclassified

An Overview of Methods for Deep Learning in Neural Networks

2017

CMSE

View full text Add to dashboard Cite

Associative Storage and Retrieval of Digital Information in Networks of Adaptive “Neurons”

Cited by 51 publications

References 0 publications

Pre-production results demonstrating multiple-system models for yield analysis

Pre-production results demonstrating multiple-system models for yield analysis

Stable task information from an unstable neural population

An Overview of Methods for Deep Learning in Neural Networks

Contact Info

Product

Resources

About