Number of stable points for spin-glasses and neural networks of higher orders.

By Pierre Baldi, Santosh S. Venkatesh. 1987

Subfields: Probabilistic & Bayesian Methods, AI for Physical Sciences

How researchers see this paper

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

In Probabilistic & Bayesian Methods

1. Jascha Narain Sohl-Dickstein
2. Dmitry Krotov
3. Viet-Hung Tran
4. David Sussillo
5. John J. Hopfield
6. Sepp Hochreiter
7. Jasmine Collins
8. E. Gardner

In AI for Physical Sciences

1. Jascha Narain Sohl-Dickstein
2. Dmitry Krotov
3. John J. Hopfield
4. Sepp Hochreiter
5. E. Gardner

Citations in the corpus

• CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP (2021, background): “Their storage capacity can be considerably increased by polynomial terms in the energy function (Chen et al., 1986; Psaltis & Cheol, 1986; Baldi & Venkatesh, 1987; Gardner, 1987; Abbott & Arian, 1987; Horn & Usher, 1988; Caputo & Niemann, 2002; Krotov & Hopfield, 2016).”
• Capacity and Trainability in Recurrent Neural Networks (2016, background): “A foundational result in machine learning is that a single-layer perceptron with N 2 parameters can store at least 2 bits of information per parameter (Cover, 1965; Gardner, 1988; Baldi & Venkatesh, 1987).”
• Dense Associative Memory for Pattern Recognition (2016, methodology): “For higher powers n the capacity rapidly grows with N in a non-linear way, allowing the network to store and reliably retrieve many more patterns than the number of neurons that it has, in accord1 with [13, 14, 15, 16].”
• History Compression via Language Models in Reinforcement Learning (2022, background): “Their storage capacity can be considerably increased by polynomial terms in the energy function (Chen et al., 1987; Psaltis & Park, 1986; Baldi & Venkatesh, 1987; Gardner, 1987; Abbott & Arian, 1987; Horn & Usher, 1988; Caputo & Niemann, 2002; Krotov & Hopfield, 2016).”
• The space of interactions in neural network models (1988, background): “Taking the limit K + 0 in equation ( 2 5 ) gives a , = 2 in agreement with the known results (Cover 1965, Venkatesh 1986a, b, Baldi and Venkatesh 1987).”