Editors' Choice

Science  12 Oct 2018:
Vol. 362, Issue 6411, pp. 195

eLetters is an online forum for ongoing peer review. Submission of eLetters are open to all. eLetters are not edited, proofread, or indexed.  Please read our Terms of Service before submitting your own eLetter.

Compose eLetter

Plain text

  • Plain text
    No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Author Information
First or given name, e.g. 'Peter'.
Your last, or family, name, e.g. 'MacMoody'.
Your email address, e.g. higgs-boson@gmail.com
Your role and/or occupation, e.g. 'Orthopedic Surgeon'.
Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.
Statement of Competing Interests
CAPTCHA

This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.

Image CAPTCHA
Enter the characters shown in the image.

Vertical Tabs

  • A Feedforward neural network is a subset of a recurrent neural network

    Peter Stern wrote a short article entitled “Memory recirculation and integration” (1). We have known that a feedforward neural network is a subset of a recurrent neural network with suppressing feedback parameters. When completely eliminating the feedback signals (no feedback) by changing synaptic links, any recurrent neural network will become a feedforward neural network. Therefore, controlling the strength of synaptic links between neurons, we may keep / forget a memory where the transition of the system state indicates a sequence of episodes. It may happen that a human-made model or a mathematical model is created before discovering the model in nature. In other words, reasoning deductive conclusion becomes always true as long as all deductive rules are true.

    References:
    1. Peter Stern, Memory recirculation and integration, Science 12 Oct 2018: Vol. 362, Issue 6411, pp. 195

    Competing Interests: None declared.