Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

Transfer between long-term and short-term memory using Conceptors

Anthony Strock 1, 2, 3 Nicolas Rougier 1, 2, 3 Xavier Hinaut 1, 2, 3
1 Mnemosyne - Mnemonic Synergy
LaBRI - Laboratoire Bordelais de Recherche en Informatique, Inria Bordeaux - Sud-Ouest, IMN - Institut des Maladies Neurodégénératives [Bordeaux]
Abstract : We introduce a recurrent neural network model of working memory combining short-term and long-term components. e short-term component is modelled using a gated reservoir model that is trained to hold a value from an input stream when a gate signal is on. e long-term component is modelled using conceptors in order to store inner temporal pa erns (that corresponds to values). We combine these two components to obtain a model where information can go from long-term memory to short-term memory and vice-versa and we show how standard operations on conceptors allow to combine long-term memories and describe their e ect on short-term memory.
Complete list of metadatas

Cited literature [25 references]  Display  Hide  Download

https://hal.inria.fr/hal-02494493
Contributor : Xavier Hinaut <>
Submitted on : Tuesday, March 10, 2020 - 4:17:50 PM
Last modification on : Monday, April 6, 2020 - 2:35:09 PM
Long-term archiving on: : Thursday, June 11, 2020 - 2:46:53 PM

Files

draft_cog_comp-without-preambu...
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02494493, version 1
  • ARXIV : 2003.11640

Citation

Anthony Strock, Nicolas Rougier, Xavier Hinaut. Transfer between long-term and short-term memory using Conceptors. 2020. ⟨hal-02494493⟩

Share

Metrics

Record views

90

Files downloads

248