Information bottleneck-based hebbian learning rule naturally ties working memory and synaptic updates

HIGHLIGHTS

• What: The work is similar in that the authors propose a dramatically different training objective. The authors explore this connection in some initial experiments on memory size and learning performance. The authors show that this global component can be computed using an auxiliary network. The authors show that optimizing the HSIC bottleneck via gradient descent emits a three-factor learning rule (Frémaux and Gerstner, 2016) composed of a local Hebbian component and a global layer-wise modulating signal.
• Who: Kyle Daruwalla from the Tsinghua University, China have published the Article: Information bottleneck-based Hebbian . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.