Toward robust and scalable deep spiking reinforcement learning

HIGHLIGHTS

• who: Mahmoud Akl from the University of Sussex, United Kingdom have published the Article: Toward robust and scalable deep spiking reinforcement learning, in the Journal: (JOURNAL)
• what: The authors explore randomizing membrane parameters across the entire network, and observe that this approach improves SNN training with surrogate gradients. This approach showed an improved classification accuracy of SNNs when measured on several benchmark datasets.
• how: Other factors like the ability to process high-dimensional data in real time particularly when data is provided from asynchronous sensors like event-based cameras (Gallego et_al 2022) provide . . .
  

   

  

  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.