Hebbian learning, its correlation catastrophe, and unlearning

How researchers see this paper

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

Catastrophic Forgetting and the Pseudorehearsal Solution in Hopfield-type Networks (1998, methodology): “This technique is then applied to both the pseudorehearsal process and the unlearning process (Hopfield, Feinstein, and Palmer, 1983; van Hemmen, 1997) in Chapter 6, to establish if performance can be improved.”
Sequential learning in neural networks: A review and a discussion of pseudorehearsal based methods (2004, unknown): “This problem has been addressed in the literature by a method known as “unlearning”, ﬁrst proposed in [10,33], and extended in other studies such as [9,67,68].”
Catastrophic forgetting in simple networks: an analysis of the pseudorehearsal solution. (1999, methodology): “§ Although not directly connected to the catastrophic learning literature, the ‘unlearning’ method [8, 3, 23, 2, 22] addresses the same issues of capacity and sequential learning tasks in Hopfield networks.”
The consolidation of learning during sleep: comparing the pseudorehearsal and unlearning accounts (1999, unknown): “Van Hemmen ( 1997 ) lets the network learn a large number of patterns before the start of unlearning , whereas Christos ( 1996 ) performs unlearning after only a few patterns have been learned .”