The consolidation of learning during sleep: comparing the pseudorehearsal and unlearning accounts

We suggest that any brain-like (artificial neural network based) learning system will need a sleep-like mechanism for consolidating newly learned information if it wishes to cope with the sequential/ongoing learning of significantly new information. We summarise and explore two possible candidates for a computational account of this consolidation process in Hopfield type networks. The "pseudorehearsal" method is based on the relearning of randomly selected attractors in the network as the new information is added from some second system. This process is supposed to reinforce old information within the network and protect it from the disruption caused by learning new inputs. The "unlearning" method is based on the unlearning of randomly selected attractors in the network after new information has already been learned. This process is supposed to locate and remove the unwanted associations between information that obscure the learned inputs. We suggest that as a computational model of sleep consolidation, the pseudorehearsal approach is better supported by the psychological, evolutionary, and neurophysiological data (in particular accounting for the role of the hippocampus in consolidation).

[1]  Eytan Ruppin,et al.  Memory Maintenance via Neuronal Regulation , 1998, Neural Computation.

[2]  Anders Krogh,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[3]  Samuel Roll,et al.  Frequency of Day Residue in Dreams of Young Adults , 1992, Perceptual and motor skills.

[4]  Robert M. French,et al.  Semi-distributed Representations and Catastrophic Forgetting in Connectionist Networks , 1992 .

[5]  D W Brown.  Crick and Mitchison's theory of REM sleep and neural networks. , 1993, Medical hypotheses.

[6]  Noel E. Sharkey,et al.  An Analysis of Catastrophic Interference , 1995, Connect. Sci..

[7]  M. A. Moore,et al.  Neural network models of list learning , 1991 .

[8]  Anthony Robins.  Maintaining stability during new learning in neural networks , 1997, 1997 IEEE International Conference on Systems, Man, and Cybernetics. Computational Cybernetics and Simulation.

[9]  James L. McClelland,et al.  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, Psychological review.

[10]  Charles J. Brainerd,et al.  New perspectives on interference and inhibition in cognition: Final comments , 1995 .

[11]  J. L. Kavanau,et al.  Memory, sleep and the evolution of mechanisms of synaptic efficacy maintenance , 1997, Neuroscience.

[12]  Anthony V. Robins,et al.  Consolidation in Neural Networks and in the Sleeping Brain , 1996, Connect. Sci..

[13]  R. French.  Dynamically constraining connectionist networks to produce distributed, orthogonal representations to reduce catastrophic interference , 2019, Proceedings of the Sixteenth Annual Conference of the Cognitive Science Society.

[14]  J. L. van Hemmen,et al.  Hebbian learning, its correlation catastrophe, and unlearning , 1997 .

[15]  Anthony V. Robins,et al.  Catastrophic Forgetting, Rehearsal and Pseudorehearsal , 1995, Connect. Sci..

[16]  K. McRae,et al.  Catastrophic Interference is Eliminated in Pretrained Networks , 1993 .

[17]  J. Winson,et al.  The meaning of dreams. , 1990, Scientific American.

[18]  Johanna D. Moore.  Proceedings of the fifteenth annual meeting of the Cognitive Science Society , 1993 .

[19]  Mitsuaki Yamamoto,et al.  Metastable associative network models of dream sleep , 1997, Neural Networks.

[20]  D. J. Wallace,et al.  Training with noise and the storage of correlated patterns in a neural network model , 1989 .

[21]  D. Signorini,et al.  Neural networks , 1995, The Lancet.

[22]  Francis Crick,et al.  REM sleep and neural nets , 1995, Behavioural Brain Research.

[23]  Marcus R. Frean,et al.  A "Thermal" Perceptron Learning Rule , 1992, Neural Computation.

[24]  J. Leo van Hemmen,et al.  Universality of unlearning , 1994, Neural Networks.

[25]  Carlyle Smith,et al.  Sleep states, memory processes and synaptic plasticity , 1996, Behavioural Brain Research.

[26]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[27]  Anthony V. Robins,et al.  Catastrophic Forgetting and the Pseudorehearsal Solution in Hopfield-type Networks , 1998, Connect. Sci..

[28]  Francis Crick,et al.  The function of dream sleep , 1983, Nature.

[29]  Geoffrey E. Hinton.  Using fast weights to deblur old memories , 1987 .

[30]  J. Stephen Judd,et al.  Learning in neural networks , 1988, COLT '88.

[31]  J. Hobson,et al.  Modeling States of Waking and Sleeping , 1992 .

[32]  M. Posner.  Foundations of cognitive science , 1989 .

[33]  B. Hars,et al.  Processing of learned information in paradoxical sleep: relevance for memory , 1995, Behavioural Brain Research.

[34]  R. Hoffmann,et al.  The Functions of Dreaming , 1993 .

[35]  Stanislas Dehaene,et al.  Networks of Formal Neurons and Memory Palimpsests , 1986 .

[36]  G Buzsáki,et al.  Memory consolidation during sleep: a neurophysiological perspective. , 1998, Journal of sleep research.

[37]  Francesco E. Lauria,et al.  On a Learning Neural Network , 1988 .

[38]  L. F. Abbott.  Learning in neural network memories , 1990 .

[39]  A. Giuditta,et al.  The sequential hypothesis of the function of sleep , 1995, Behavioural Brain Research.

[40]  Michael McCloskey,et al.  Catastrophic Interference in Connectionist Networks: The Sequential Learning Problem , 1989 .

[41]  B. McNaughton,et al.  Reactivation of hippocampal ensemble memories during sleep. , 1994, Science.

[42]  S. Lewandowsky,et al.  Catastrophic interference in neural networks , 1995 .

[43]  Geoffrey E. Hinton,et al.  Learning and relearning in Boltzmann machines , 1986 .

[44]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[45]  Jacob M.J. Murre,et al.  Learning and Categorization in Modular Neural Networks , 1992 .

[46]  Robert M. French,et al.  Pseudo-recurrent Connectionist Networks: An Approach to the 'Sensitivity-Stability' Dilemma , 1997, Connect. Sci..

[47]  Stephen Grossberg,et al.  Competitive Learning: From Interactive Activation to Adaptive Resonance , 1987, Cogn. Sci..

[48]  R. Greenberg,et al.  Cutting the REM Nerve: An Approach to the Adaptive Role of REM Sleep , 2015, Perspectives in biology and medicine.

[49]  B D Davis,et al.  Sleep and the Maintenance of Memory , 2015, Perspectives in biology and medicine.

[50]  J. J. Hopfield,et al.  ‘Unlearning’ has a stabilizing effect in collective memories , 1983, Nature.

[51]  David E. Rumelhart,et al.  The architecture of mind: a connectionist approach , 1989 .

[52]  Adam N. Mamelak,et al.  Models Wanted: Must Fit Dimensions of Sleep and Dreaming , 1991, NIPS.

[53]  R Ratcliff,et al.  Connectionist models of recognition memory: constraints imposed by learning and forgetting functions. , 1990, Psychological review.

[54]  George A. Christos,et al.  Investigation of the Crick-Mitchison reverse-learning dream sleep hypothesis in a dynamical setting , 1996, Neural Networks.