Use of Magnetoresistive Random-Access Memory as Approximate Memory for Training Neural Networks

25 Oct 2018 Locatelli Nicolas Vincent Adrien F. Querlioz Damien

Hardware neural networks that implement synaptic weights with embedded non-volatile memory, such as spin torque memory (ST-MRAM), are a major lead for low energy artificial intelligence. In this work, we propose an approximate storage approach for their memory... (read more)

PDF Abstract
No code implementations yet. Submit your code now

Categories


  • EMERGING TECHNOLOGIES
  • APPLIED PHYSICS