If Neural Networks Are Allowed To Sleep And Dream, Their Performance Sensibly Increases

The harmonic oscillator for associative memory and pattern recognition in Artificial Intelligence is certainly the Hopfield model [1] (or, equivalently [2], its dual representation, i.e. the Restricted Boltzmann Machine (RBM) [3]). In a nutshell, we can store information (consisting in a set of P digital words or -generally speaking- patterns of information) by suitably modifying the synaptic interactions among neurons in the Hopfield neural network by means of the so-called Hebbian learning (or by using contrastive divergence algorithms for training RBMs).

However, since the seminal AGS theory [5], we know that -for this model- the critical capacity αc (namely the maximal amount of patterns Pmax a network equipped with N neurons can handle, i.e. αc = Pmax/N), is αc ∼ 0.14. On the other side, general information theoretical considerations (i.e., Gardner Theory [4]) prescribe that the maximal critical capacity for networks equipped with symmetric couplings (as the Hopfield model) is precisely αc = 1 (i.e. Pmax = N), significantly better than the Hopfield case.


Recently, in our work [6], we have shown that forcing the network to sleep -namely, mathematically implementing dreaming processes (mimicking slow wave sleep (SWS) and random eye movements (REM) phases in real mammal’s brain), after a proper rest the Hopfield model has impressive enhanced skills. In particular, it reaches the maximal bound for the critical storage, i.e. αc = 1, further, stored patterns are very stable configurations (technically speaking, they are free energy absolute minima) up to α ∼ 0.85 (while in standard Hopfield picture, this is true solely for α < 0.05). This is a remarkable property as, in the long time limit, any stochastic neural dynamics for the standard Hopfield model, if α > 0.05, almost surely ends up in a (spin-glass) state that does not overlap significantly with any of the stored patterns; while -if the network can sleep- this happens only for α > 0.85 conferring to network’s functionalities an enhanced robustness with respect to severe loads of patterns.

Finally, thanks to the equivalence between RBMs and Hopfield networks [2], we know that the ratio α = P/N in the Hopfield retrieval matches the ratio among the size of the hidden layer (built of P hidden neurons) over that of the visible one (build of N visible neurons) for RBMs. In the standard scenario, if the hidden layer is larger than ∼ 0.14 times the visible one, learning can be prone to over-fitting (as its dual Hopfield model would lie already outside the working regime, in the spin-glass phase). Allowing the network to sleep, we can learn without over-fitting even by keeping the two layers roughly of the same size (and this obviously allows to extract much more features from the data-sets presented to the network during the training stage).

Summarizing, we believe that in the process of cognition -while certainly learning and retrieval keep covering a pivotal role – also sleeping is mandatory for Artificial Intelligence, as it is for the biological one.

These findings are described in the article entitled Dreaming neural networks: Forgetting spurious memories and reinforcing pure ones, recently published in the journal Neural Networks.



  1. J.J. Hopfield, Neural networks and physical systems with emergent collective computational abilities, Proc. Natl. Acad. Sci. 79(8):2554, (1982).
  2. A. Barra, et al., On the equivalence of Hopfield networks and Boltzmann machines, Neural Networks 34:1, (2012).
  3. D.H. Ackley, G.E. Hinton, T.J. Sejnowski, A learning algorithm for Boltzmann machines, Cognitive science 9(1):147, (1985).
  4. E. Gardner, The space of interactions in neural network models, J. Phys. A 21(1):257, (1998).
  5. D.J. Amit, H. Gutfreund, H. Sompolinsky, Storing infinite numbers of patterns in a spin-glass model of neural networks, Phys. Rev. Lett. 55(14):1530, (1985).
  6. A. Fachechi, E. Agliari, A. Barra, Dreaming neural networks: forgetting spurious memories and reinforcing pure ones, Neural Networks (2019) in press



Do Sports Fans Connect More With Endangered Species Mascots?

Sports team mascots serve numerous purposes for sports organizations within the United States.  Mascots provide a point of attachment for […]

Tropical Savanna: Animals Plants & Climate

A tropical savanna is an environment characterized by rolling hills, tall grasslands, and sparse trees. Large parts of Africa and […]

What’s The Unintended Consequence Of Electric Cars?

Ozone. No, not the ‘good’ ozone up in the stratosphere that protects us from cosmic rays. I mean the ‘bad’ […]

Yolk / Shell Nanocrystals: A Novel Approach To Catalysis and Drug Release

Yolk/shell (Y–S) hybrid nanocrystals (HNCs) has become an emerging hot-spot material towards space-confined nanocatalysis, nanoreactors, drug release, and energy conversion, […]

The Backside Of Paintings: Why It’s Time For The Back Of Paintings To Get A Regeneration Process

When looking at paintings at the museum, the observer’s attention is literally captured by the colors, the forms, the feelings […]

7 Continents And 5 Oceans

The 7 continents of the world are North America, South America, Antarctica, Europe, Asia, Africa, and Australia. The 5 oceans […]

A New Cost-Effective Deployment Strategy For Satellite Constellations

Published by Hang Woon Lee and Koki Ho Department of Aerospace Engineering, University of Illinois at Urbana-Champaign These findings are […]

Science Trends is a popular source of science news and education around the world. We cover everything from solar power cell technology to climate change to cancer research. We help hundreds of thousands of people every month learn about the world we live in and the latest scientific breakthroughs. Want to know more?