Daydreaming Hopfield Networks and their surprising effectiveness on correlated data
arXiv:2405.08777 · doi:10.1016/j.neunet.2025.107216
Abstract
To improve the storage capacity of the Hopfield model, we develop a version of the dreaming algorithm that perpetually reinforces the patterns to be stored (as in the Hebb rule), and erases the spurious memories (as in dreaming algorithms). For this reason, we called it Daydreaming. Daydreaming is not destructive and it converges asymptotically to stationary retrieval maps. When trained on random uncorrelated examples, the model shows optimal performance in terms of the size of the basins of attraction of stored examples and the quality of reconstruction. We also train the Daydreaming algorithm on correlated data obtained via the random-features model and argue that it spontaneously exploits the correlations thus increasing even further the storage capacity and the size of the basins of attraction. Moreover, the Daydreaming algorithm is also able to stabilize the features hidden in the data. Finally, we test Daydreaming on the MNIST dataset and show that it still works surprisingly well, producing attractors that are close to unseen examples and class prototypes.
References in corpus (8)
- Supervised Hebbian Learning
- The Exponential Capacity of Dense Associative Memories
- Equilibrium and non-Equilibrium regimes in the learning of Restricted Boltzmann Machines
- Learning through atypical "phase transitions" in overparameterized neural networks
- Supervised perceptron learning vs unsupervised Hebbian unlearning: Approaching optimal memory retrieval in Hopfield-like networks
- Probing transfer learning with a model of synthetic correlated datasets
- Dense Hebbian neural networks: a replica symmetric picture of supervised learning
- Random Features Hopfield Networks generalize retrieval to previously unseen examples