I am trying to understand how far associative networks have evolved from the Hopfield network. A lot of the research is only available to institutions which is why I deferred to this stack exchange. My belief is that modern networks have fully connected networks with a basic perceptron design on the inside with weights and biases. The inputs consist of the state of its neighboring nodes. Is there any truth to my understanding or is there anything I am missing? Could you please refer me to some places to better understand the modern state of associative memory?


I am referencing the modern versions of the Hopfield network described in this paper: https://www.pnas.org/doi/epdf/10.1073/pnas.79.8.2554

I am not referencing any particular paper but merely want examples of modern DANs in an attempt to see how much they have changed from the original one.


If this is not the right place to ask this question, could you please refer me to right place to get my answer? I truly appreciate all your help!

  • $\begingroup$ Could you add some references? Also, are you referring to some paper in particular? $\endgroup$ May 31, 2023 at 6:54
  • $\begingroup$ I edited my question. Please let me know if you need any more information. $\endgroup$ May 31, 2023 at 18:43
  • $\begingroup$ Can you please reformulate the title as a specific question? Thanks. $\endgroup$
    – nbro
    Jun 1, 2023 at 8:33


You must log in to answer this question.

Browse other questions tagged .