# Adaptive Precision CNN Accelerator Using Radix-X Parallel Connected Memristor Crossbars

@article{Lee2019AdaptivePC, title={Adaptive Precision CNN Accelerator Using Radix-X Parallel Connected Memristor Crossbars}, author={Jaeheum Lee and Jason K. Eshraghian and Kyoung-Rok Cho and Kamran Eshraghian}, journal={ArXiv}, year={2019}, volume={abs/1906.09395} }

Neural processor development is reducing our reliance on remote server access to process deep learning operations in an increasingly edge-driven world. By employing in-memory processing, parallelization techniques, and algorithm-hardware co-design, memristor crossbar arrays are known to efficiently compute large scale matrix-vector multiplications. However, state-of-the-art implementations of negative weights require duplicative column wires, and high precision weights using single-bit… CONTINUE READING

#### Citations

##### Publications citing this paper.

## Reconfigurable multiplier architecture based on memristor-cmos with higher flexibility

VIEW 1 EXCERPT

CITES METHODS

#### References

##### Publications referenced by this paper.

SHOWING 1-10 OF 39 REFERENCES

## Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1

VIEW 10 EXCERPTS

HIGHLY INFLUENTIAL

## Analog Weights in ReRAM DNN Accelerators

VIEW 1 EXCERPT

## Deep Learning towards Mobile Applications

VIEW 1 EXCERPT