Mitigating Adversarial Attack for Compute-in-Memory Accelerator Utilizing On-chip Finetune.
Shanshi HuangHongwu JiangShimeng YuPublished in: NVMSA (2021)
Keyphrases
- compute intensive
- high speed
- random access memory
- memory requirements
- memory access
- low cost
- parallel implementation
- vlsi implementation
- memory subsystem
- computing power
- level parallelism
- multithreading
- data corruption
- digital signal processors
- countermeasures
- high density
- dynamic random access memory
- neural network
- programmable logic
- memory size
- high bandwidth
- ddos attacks
- memory usage
- data structure