Gene/Protein Disease Symptom Drug Enzyme Compound
Pivot Concepts:   Target Concepts:
Query: UMLS:C0598853 (forgetting)
 3,232 document(s) hit in 31,850,051 MEDLINE articles (0.00 seconds)

We propose learning and forgetting techniques for the generalized brain-state-in-a-box (BSB) based associative memories. A generalization of the BSB model allows each neuron to have its own bias and the synaptic weight matrix does not have to be symmetric. A pattern is learned by a memory if its noisy or an incomplete version presented to the memory is mapped back to this pattern. A pattern, previously stored, is forgotten or deleted from the memory if a stimulus that is a perturbed version of the pattern, when presented to the memory, is not mapped back to this pattern. In this paper we propose "on-line" memory storage and deletion methods using an iterative method of computing the pseudo-inverse of a given matrix. The proposed methods allow one to "add" or "delete" a memory pattern by updating, rather than recomputing from scratch, the current synaptic weight matrix in a single step. We first analyze the desired characteristics of neural network associative memories. After that, we review the existing methods for design of neural associative memories. Then we discuss the generalized BSB neural model and its possible function as an associative memory and proffer arguments in support of using such models for neural associative memories. In particular, the generalized BSB type models are easier to analyze, synthesize, and implement than other neural networks. The results obtained are illustrated by numerical examples. Copyright 1996 Elsevier Science Ltd
 ...
PMID:Learning and Forgetting in Generalized Brain-state-in-a-box (BSB) Neural Associative Memories. 1266 67

Many companies assume that once they've launched a major innovation, growth will soon follow. It's not that simple. High-potential new businesses within established companies face stiff headwinds well after their inception. That's why a company's emphasis must shift: from ideas to execution and from leadership excellence to organizational excellence. The authors spent five years chronicling new businesses at the New York Times Company, Analog Devices, Corning, Hasbro, and other organizations. They found that a breakthrough new business (referred to as NewCo) rarely coexists gracefully with the established business in the company (called CoreCo). The unnatural combination creates three specific challenges--forgetting, borrowing, and learning--that NewCo must meet in order to survive and grow. NewCo must first forget some of what made CoreCo successful. NewCo and CoreCo have elemental differences, so NewCo must leave behind CoreCo's notions about what skills and competencies are most valuable. NewCo must also borrow some of CoreCo's assets--usually in one or two key areas that will give NewCo a crucial competitive advantage. Incremental cost reductions, for example, are never a sufficient justification for borrowing. Finally, NewCo must be prepared to learn some things from scratch. Because strategic experiments are highly uncertain endeavors, NewCo will face several critical unknowns. The more rapidly it can resolve those unknowns--that is, the faster it can learn--the sooner it will zero in on a winning business model or exit a hopeless situation. Managers can accelerate this learning by planning more simply and more often and by comparing predicted and actual trends.
 ...
PMID:Building breakthrough businesses within established organizations. 1592 4