Learning and connectionist representations

By David E. Rumelhart, Peter M. Todd. 1993

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

A mathematical theory of semantic development in deep neural networks (2018, background): “How could such changing patterns of inductive generalization arise in a neural network? Here, building upon previous network simulations of inductive projection [4, 31], we show analytically that deep networks exposed to hierarchically structured data, naturally yield progressively narrowing pattern”
Précis of Semantic Cognition: A Parallel Distributed Processing Approach (2008, methodology): “As previously mentioned, the models we will use to illustrate the theory are variants of an architecture first proposed by Rumelhart (Rumelhart 1990; Rumelhart & Todd 1993) as illustrated in Figure 1.”
Learning hierarchical categories in deep neural networks (2013, methodology): “To begin, though, we consider the case of orthogonal input representations where each item is designated by a single active input unit, as used by (Rumelhart & Todd, 1993) and (Rogers & McClelland, 2004).”
Semantic Cognition: A Parallel Distributed Processing Approach (2004, methodology): “The Rumelhart Feed-Forward Network . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .”
Integration of New Information in Memory: New Insights from a Complementary Learning Systems Perspective (2020, methodology): “The network (shown in Figure 2) was initially conceived by Rumelhart & Todd [5] as a way of learning knowledge that might be expressed through propositional statements such as ’a canary can ﬂy’, ’a tree is a plant’, ’a rose is red’ and ’ﬁsh have ﬁns’.”
Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. (1995, methodology): “Lesions to the hippocampus proper, sparing the structures in the parahippocampal region, can produce quantitatively smaller de cits, or de cits that show up only on certain tasks and or at long delay (see Squire, 1992; Eichenbaum et al., in press for reviews). We discuss this issue further in the Ge”
Connectionist models of development (2003, methodology): “Following the work of Rumelhart and Todd (1993) , Rogers and McClelland (in press; McClelland & Rogers, 2003) considered a network learning about the properties of various specific objects chosen from the categories of birds, fish, trees and flowers.”
Learning hierarchical category structure in deep neural networks (2013, methodology): “Many neural network simulations have captured aspects of these broad patterns of semantic development (Rogers & Mc- Clelland, 2004; Rumelhart & Todd, 1993; McClelland, 1995; Plunkett & Sinha, 1992; Quinn & Johnson, 1997).”
Structure and deterioration of semantic memory: a neuropsychological and computational investigation. (2004, background): “Items that share few verbal descriptors and have different visual appearances contribute different inputs and targets to the system, and they generate differing internal representations ( Rumelhart & Todd, 1993; Plaut, 2002; McClelland & Rogers, 2003; Rogers & McClelland, in press).”
Situated simulation in the human conceptual system (2003, background): “For specific applications of the feed-forward architecture to representing conceptual knowledge, see Hinton (1989), Kruschke (1992), and Rumelhart and Todd (1993).”