Solving the spike feature information vanishing problem in spiking deep q network with potential based normalization

HIGHLIGHTS

• who: Yi Zeng from the Institute of Automation (CAS), China University of Massachusetts Amherst have published the research work: Solving the spike feature information vanishing problem in spiking deep Q network with potential based normalization, in the Journal: (JOURNAL)
• what: The authors mathematically analyzed the problem of the disappearance of spiking signal features in SDQN and proposed a potential-based layer normalization (pbLN) method to train spiking deep Q networks directly. The spiking deep Q network is directly trained with a surrogate function, and the experiments show that the pbLN improves the performance of SNN . . .
  

   

  

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

  Thanks :)

  Username or E-mail:

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