An Analysis of Catastrophic Interference

By Noel E. Sharkey, Amanda J. C. Sharkey. 1995

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

Measuring Catastrophic Forgetting in Neural Networks (2017, background): “Sparse representations can reduce the chance of this interference; however, sparsity can impair generalization and ability to learn new tasks (Sharkey and Sharkey 1995).”
Amnesiac Machine Learning (2020, background): “The principle of catastrophic interference (Sharkey and Sharkey 1995) tells us that when presented with new data, there is some probability of an artiﬁcial neural network losing previously learned information.”
Sequential learning in neural networks: A review and a discussion of pseudorehearsal based methods (2004, unknown): “As CF has been well discussed in the literature we provide only a brief introduction here, see also [51, 63], or see [21] for a recent and comprehensive overview.”
Catastrophic Forgetting and the Pseudorehearsal Solution in Hopfield-type Networks (1998, background): “A number of studies have used MLP type networks (typically using back-propagation for training) to highlight the problem of CF and explore various issues (McCloskey and Cohen, 1989; Hetherington and Seidenberg, 1989; Ratcliff, 1990; Lewandowsky, 1991; Murre, 1992; French, 1992, 1994, 1997; McRae and”
The Primacy Bias in Deep Reinforcement Learning (2022, result): “Forgetting mechanisms In contrast to the well-known phenomenon of catastrophic forgetting (French, 1999), several works have noted the opposite effect of catastrophic memorization (Robins, 1993; Sharkey & Sharkey, 1995) similar to the primacy bias.”
Hybrid neural systems: from simple coupling to fully integrated neural networks (1999, background): “Adding extra training examples may cause problems unless re-training is carried out using the original training set along with the new examples, otherwise \catastrophic interference" may occur [78].”
Incremental learning algorithms and applications (2016, background): “Indeed, localist connectionist models such as radial basis function (RBF) networks can work reliably in incremental settings [133, 100], whereby care has to be taken to guarantee their generalisation performance [147].”
Incremental learning with self-organizing maps (2017, background): “As the problem of catastrophic forgetting was first remarked for multilayer perceptron (MLP) models [4], [5], it is hardly surprising that there is significant work on the subject of how catastrophic forgetting could be avoided[16], [17], [18], [19], [20], [21], [22], [23] although none of these pro”
Connectionist modelling in psychology: A localist manifesto (2000, methodology): “This is, of course, exactly what is achieved by using localist representations as intermediates (for a review and a look-up model similar to that proposed here, see Sharkey & Sharkey 1995).”
Pseudo-recurrent Connectionist Networks: An Approach to the 'Sensitivity-Stability' Dilemma (1997, methodology): “One particular problem that this model cannot overcome is, ironically, the flip side of catastrophic forgetting, what Sharkey & Sharkey (1995) have called “catastrophic remembering.”.”
Critical periods and catastrophic interference effects in the development of self-organizing feature maps. (2008, unknown): “…typically qualitative in nature (e.g. catastrophic interference effects: French & Chater, 2002; McCloskey & Cohen, 1989; Ratcliff, 1990; Robins, 1995; Robins & McCallum, 1998; Sharkey & Sharkey, 1995; critical period effects: McClelland et al., 1999; Oliver et al., 2000; O’Reilly & Johnson, 1994).”