Realizing Forgetting in a Modified Sparse Distributed Memory System
نویسندگان
چکیده
This paper presents research on the development of effective forgetting mechanisms for the Sparse Distributed Memory (SDM) system, to computationally model Transient Episodic Memory (TEM), a short-term sensory perceptual episodic memory in software agents. Possible theories and mechanisms for forgetting are retrieval failures, decay and interference. The SDM architecture has inherent features to effect interference and retrieval failures. We have implemented two decay mechanisms in a variant of the SDM system. In this paper, we present the decay mechanisms and the experimental results. The results show that the decay mechanisms compliment the inherent features of the SDM architecture in realizing forgetting for TEM.
منابع مشابه
Integer Sparse Distributed Memory
Sparse distributed memory is an auto associative memory system that stores high dimensional Boolean vectors. Here we present an extension of the original SDM, the Integer SDM that uses modular arithmetic integer vectors rather than binary vectors. This extension preserves many of the desirable properties of the original SDM: auto associativity, content addressability, distributed storage, and r...
متن کاملEncoding and Retrieval Efficiency of Episodic Data in a Modified Sparse Distributed Memory System
This paper presents detailed simulation results on a modified Sparse Distributed Memory (SDM) system. We have modified Kanerva’s original SDM system into an architecture with a ternary memory space. This enables the memory to be used as a Transient Episodic Memory (TEM) in cognitive software agents. TEM is a memory with high specificity and low retention, used for events having features of a pa...
متن کاملModeling 1D Distributed-Memory Dense Kernels for an Asynchronous Multifrontal Sparse Solver
To solve sparse linear systems multifrontal methods rely on dense partial LU decompositions of so-called frontal matrices; we consider a parallel, asynchronous setting in which several frontal matrices can be factored simultaneously. In this context, to address performance and scalability issues of acyclic pipelined asynchronous factorization kernels, we study models to revisit properties of le...
متن کاملSparse distributed memory using N-of-M codes
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 net...
متن کاملUsing the Modified Shuffled Frog Leaping Algorithm for Optimal Sizing and location of Distributed Generation Resources for Reliability Improvement
Restructuring the recent developments in the power system and problems arising from construction as well as the maintenance of large power plants lead to increase in using the Distributed Generation (DG) resources. DG units due to its specifications, technology and location network connectivity can improve system and load point reliability indices. In this paper, the allocation and sizing of di...
متن کامل