A neurodynamical model for working memory

By Razvan Pascanu, Herbert Jaeger. 2011

85 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Electrical and Electronic Engineering, 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 Artificial Intelligence

1. Yoshua Bengio
2. Razvan Pascanu
3. Kyunghyun Cho
4. J. Urgen Schmidhuber
5. Mohit Bansal
6. Quoc V. Le
7. Jianfeng Gao
8. Ruslan Salakhutdinov

In Cognitive Neuroscience

1. Yoshua Bengio
2. Jean-Pierre Changeux
3. Stanislas Dehaene
4. Catherine Wacongne
5. Razvan Pascanu
6. Herbert Jaeger
7. Seiji Takeda
8. Stefan J. Kiebel

In Electrical and Electronic Engineering

1. Yoshua Bengio
2. Razvan Pascanu
3. Jean-Pierre Changeux
4. Stanislas Dehaene
5. Catherine Wacongne
6. Tomas Mikolov
7. Seiji Takeda
8. Gouhei Tanaka

In Neuroscience

1. Stanislas Dehaene
2. Catherine Wacongne
3. Jean-Pierre Changeux
4. Herbert Jaeger
5. Yoshua Bengio
6. Stefan J. Kiebel

In Cognitive Science & Psychology

1. Stanislas Dehaene
2. Catherine Wacongne
3. Jean-Pierre Changeux
4. Herbert Jaeger
5. Yoshua Bengio
6. Razvan Pascanu
7. Tomas Mikolov
8. Christopher D. Manning

Citations in the corpus

• On the difficulty of training recurrent neural networks (2012, background): “It has been shown that in practice it can reduce the chance that gradients explode, and even allow training generator models or models that work with unbounded amounts of memory(Pascanu and Jaeger, 2011; Doya and Yoshizawa, 1991).”
• Understanding the exploding gradient problem (2012, background): “It has been showed that in practice it can reduce the chance that gradients explode, and even allow training generator models or models that work with unbounded amounts of memory(Pascanu and Jaeger, 2011; Doya and Yoshizawa, 1991).”
• Re-visiting the echo state property (2012, background): “In fact, there are tasks like the multistable switching circuits described in [9] (subsequently employed to model stable working memory mechanisms [22]) which require fast reaction times of reservoirs and fast locking into attractor states.”
• Recent Advances in Physical Reservoir Computing: A Review (2018, background): “, 2007), the ESNbased working memory model (Pascanu and Jaeger, 2011), the reservoir model for storing different time constants of memory traces for reward expectation in reinforcement learning (Bernacchia et al.”
• A Neuronal Model of Predictive Coding Accounting for the Mismatch Negativity (2012, methodology): “In a more realistic memory network, the artificial delay lines that we used could be replaced by more realistic chaotic temporal dynamics, as in “reservoir” or echo state networks models (Maass et al., 2002; Buonomano, 2005; Buonomano and Laje, 2010; Pascanu and Jaeger, 2011).”