Implementing Binarized Neural Networks with Magnetoresistive RAM without Error Correction

by   Tifenn Hirtzlin, et al.

One of the most exciting applications of Spin Torque Magnetoresistive Random Access Memory (ST-MRAM) is the in-memory implementation of deep neural networks, which could allow improving the energy efficiency of Artificial Intelligence by orders of magnitude with regards to its implementation on computers and graphics cards. In particular, ST-MRAM could be ideal for implementing Binarized Neural Networks (BNNs), a type of deep neural networks discovered in 2016, which can achieve state-of-the-art performance with a highly reduced memory footprint with regards to conventional artificial intelligence approaches. The challenge of ST-MRAM, however, is that it is prone to write errors and usually requires the use of error correction. In this work, we show that these bit errors can be tolerated by BNNs to an outstanding level, based on examples of image recognition tasks (MNIST, CIFAR-10 and ImageNet): bit error rates of ST-MRAM up to 0.1 accuracy. The requirements for ST-MRAM are therefore considerably relaxed for BNNs with regards to traditional applications. By consequence, we show that for BNNs, ST-MRAMs can be programmed with weak (low-energy) programming conditions, without error correcting codes. We show that this result can allow the use of low energy and low area ST-MRAM cells, and show that the energy savings at the system level can reach a factor two.


page 1

page 2


Implementation of Fast and Power Efficient SEC-DAEC and SEC-DAEC-TAEC Codecs on FPGA

The reliability of memory devices is affected by radiation induced soft ...

Error Correction for Partially Stuck Memory Cells

We present code constructions for masking u partially stuck memory cells...

Automorphism Ensemble Polar Code Decoders for 6G URLLC

The URLLC scenario in the upcoming 6G standard requires low latency and ...

Exploration and Analysis of Combinations of Hamming Codes in 32-bit Memories

Reducing the threshold voltage of electronic devices increases their sen...

Secure synchronization of artificial neural networks used to correct errors in quantum cryptography

Quantum cryptography can provide a very high level of data security. How...

Real-time error correction and performance aid for MIDI instruments

Making a slight mistake during live music performance can easily be spot...

Memory Based Machine Intelligence Techniques in VLSI hardware

We briefly introduce the memory based approaches to emulate machine inte...

Please sign up or login with your details

Forgot password? Click here to reset