Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 15541946

Sparse distributed memory using N-of-M codes.

Steve B Furber·W John Bainbridge·J Mike Cumpstey·Steve Temple
View original source ↗Export metadata ↓

Abstract

An analysis is presented of a sparse distributed memory (SDM) inspired by that described by Kanerva [Kanerva, P. (1988). Sparse distributed memory. Cambridge, MA: MIT Press] but modified to facilitate an implementation based on spiking neurons. The memory presented here employs sparse binary N-of-M codes, unipolar binary synaptic weights and a simple Hebbian learning rule. It is a two-layer network, the first (fixed) layer being similar to the 'address decoder' in Jaeckel's [Jaeckel, L.A. (1989). A class of designs for a sparse distributed memory. RIACS Technical Report 89.30, NASA Ames Research Centre] 'hyperplane' variant of Kanerva's SDM and the second (writeable) 'data store' layer being a correlation matrix memory as first proposed by Willshaw et al. [Willshaw, D. J., Buneman, O.P., & Longuet-Higgins, H.C. (1969). Non-holographic associative memory. Nature, 222, 960-962]. The resulting network is shown to have good storage efficiency and is scalable. The analysis is supported by numerical simulations and gives results that enable the configuration of the memory to be optimised for a range of noiseless and noisy environments.

Explore related subjects

Association LearningComputer SimulationMemoryNeural Networks, ComputerJournal ArticleResearch Support, Non-U.S. Gov't

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Steve B Furber, W John Bainbridge, J Mike Cumpstey, Steve Temple. 2004. Sparse distributed memory using N-of-M codes.. https://doi.org/10.1016/j.neunet.2004.07.003

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2005

The neural effect of stimulus-response modality compatibility on dual-task performance: an fMRI study.

Recent fMRI studies suggest that the inferior frontal sulcus (IFS) is involved in the coordination of interfering processes in dual-task situations. The present study aims to further specify this assumption by investigating whether the compatibility between stimulus and response modalities modulates dual-task-related activity along the IFS. It has been shown behaviorally that the degree of interference, as measured by dual-task costs, increases in modality-incompatible conditions (e.g. visual-vocal tasks combined with auditory-manual tasks) as compared to modality-compatible conditions (e.g. visual-manual tasks combined with auditory-vocal tasks). Using fMRI, we measured IFS activity when participants performed modality-compatible and modality-incompatible single and dual tasks. Behaviorally, we replicated the finding of higher dual-task costs for modality-incompatible tasks compared to modality-compatible tasks. The fMRI data revealed higher activity along the IFS in modality-incompatible dual tasks compared with modality-compatible dual tasks when inter-individual variability in functional brain organization is taken into account. We argue that in addition to temporal order coordination (Szameitat et al., 2002), the IFS is involved in the coordination of cognitive processes associated with the concurrent mapping of sensory information onto corresponding motor responses in dual-task situations.

Association Learning↗
PubMed2005

Implicit associative learning engages the hippocampus and interacts with explicit associative learning.

The hippocampus is crucial for conscious, explicit memory, but whether it is also involved in nonconscious, implicit memory is uncertain. We investigated with functional magnetic resonance imaging whether implicit learning engages the hippocampus and interacts with subsequent explicit learning. The presentation of subliminal faces-written profession pairs for implicit learning was followed by the explicit learning of supraliminal pairs composed of the same faces combined with written professions semantically incongruous to those presented subliminally (experiment 1), semantically congruous professions (experiment 2), or identical professions (experiment 3). We found that implicit face-profession learning interacted with explicit face-profession learning in all experiments, impairing the explicit retrieval of the associations. Hippocampal activity increased during the subliminal presentation of face-profession pairs versus face-nonword pairs and correlated with the later impairment of explicit retrieval. These findings suggest that implicit semantic associative learning engages the hippocampus and influences explicit memory.

Association Learning↗
PubMed2005

Feed-forward associative learning for volitional movement control.

One of the most difficult problems in motor learning is determining the source of a learning signal, sometimes called an error signal. This problem is hidden in the adaptations of simple reflexive movements by attributing its source to sensory organs. The feed-forward associative motor learning theory proposed here attributes the source to the movement system itself. When a subject performs a corrective movement after his primary movement, the proposed neural learning device learns to associate the primary motor command with the corrective motor command by using a place-coding system. In the subsequent trials, the primary movement will involve a correction due to the participation of this mechanism, thus resulting in better performance. The theory assumes three conditions, namely, that a motor center and the learning device share the same place-encoded motor information; the motor center issues a command and a learning signal simultaneously from the same unit; and a learning signal issued with a corrective command has a heterosynaptic interaction with the previous primary command. The cerebellum is a reasonable candidate for the device satisfying these conditions. The reaction time of a corrective movement, usually 100-300 ms, almost satisfies the coincidence condition for long-term depression of the granule-to-Purkinje synapses. As an application, this theory is demonstrated to account for behavioral results regarding saccadic adaptation.

Association Learning↗