Results 281 to 290 of about 395,648 (314)
Some of the next articles are maybe not open access.
Hardware aspects of Long Short Term Memory
2018 25th IEEE International Conference on Electronics, Circuits and Systems (ICECS), 2018This paper focuses on hardware implementation aspects of a Long-Short Term Memory (LSTM) network application. Initially, certain piecewise approximations to activation functions $\sigma$ and tanh are proposed. Next a training procedure is introduced which exploits proposed piecewise approximations.
Ioannis Kouretas, Vassilis Paliouras
openaire +1 more source
Two circuits to convert short-term memory into long-term memory
Medical Hypotheses, 1997According to a brain model, encoding synapses record presynaptic axonal 'on-off' patterns as memory, and modulating synapses convert short-term memory into long-term memory by helping encoding synapses develop long-term potentiation and depression. Sensory organs conduct sounds and images as series of axonal 'on-off' patterns to encoding synapses of ...
openaire +2 more sources
A statistical theory of short and long term memory
Behavioral Biology, 1975We present a theory of short, intermediate and long term memory of a neural network incorporating the known statistical nature of chemical transmission at the synapses. Correlated pre- and post-synaptic facilitation (related to Hebb's Hypothesis) on three time scales are crucial to the model.
W A, Little, G L, Shaw
openaire +2 more sources
1990
Memory dysfunction is extremely complex, due to the variety of symptoms that occur, changes over time, and perhaps a dependency of symptoms upon location of lesion. In this chapter, adaptive problems created by memory deficits will be emphasized, i.e., frequently observed syndromes and symptoms, and some descriptive concepts that will be useful in ...
openaire +1 more source
Memory dysfunction is extremely complex, due to the variety of symptoms that occur, changes over time, and perhaps a dependency of symptoms upon location of lesion. In this chapter, adaptive problems created by memory deficits will be emphasized, i.e., frequently observed syndromes and symptoms, and some descriptive concepts that will be useful in ...
openaire +1 more source
2012
As discussed in the previous chapter, an important benefit of recurrent neural networks is their ability to use contextual information when mapping between input and output sequences. Unfortunately, for standard RNN architectures, the range of context that can be in practice accessed is quite limited.
openaire +1 more source
As discussed in the previous chapter, an important benefit of recurrent neural networks is their ability to use contextual information when mapping between input and output sequences. Unfortunately, for standard RNN architectures, the range of context that can be in practice accessed is quite limited.
openaire +1 more source
Distributed Memory for Both Short and Long Term
Neurobiology of Learning and Memory, 1998Neuropsychology points to the wide distribution of cortical memory networks. Electrophysiology and neuroimaging indicate that working memory, like long-term memory, is a widely distributed function, largely neocortical. Most of the evidence available from those three methodologies suggests that both working memory and long-term memory share the same ...
openaire +2 more sources
INTERACTION OF LONG TERM AND SHORT TERM MEMORY
The Journal of NERVOUS and MENTAL DISEASE, 1968openaire +2 more sources
A survey on long short-term memory networks for time series prediction
Procedia CIRP, 2021Benjamin Lindemann +2 more
exaly
Cell-expanded Long Short-term Memory
2022 Joint 12th International Conference on Soft Computing and Intelligent Systems and 23rd International Symposium on Advanced Intelligent Systems (SCIS&ISIS), 2022Jun Rokui, Rin Adachi
openaire +1 more source