A. Giray Yaglikci

TL;DR: EDEN is the first general framework that reduces DNN energy consumption and DNN evaluation latency by using approximate DRAM devices, while strictly meeting a user-specified target DNN accuracy, and reliably improves the error resiliency of the DNN by an order of magnitude.

TL;DR: This work proposes Copy-Row DRAM (CROW), a flexible substrate that enables new mechanisms for improving DRAM performance, energy efficiency, and reliability and uses the CROW substrate to implement a low-cost in-DRAM caching mechanism that lowers DRAM activation latency to frequently-accessed rows by 38% and a mechanism that avoids the use of short-retention-time rows to mitigate the performance and energy overhead of DRAM refresh operations.

TL;DR: In this paper, the authors present an experimental characterization of RowHammer on 1580 DRAM chips (408$\times$DDR3, 652$ \times/$DDR4, and 520$\ts$LPDDR4) from 300 DRAM modules (60$\ times$DRA3, 110$\te$DDA3, and 130$ \te$LDA3) with protection mechanisms disabled, spanning multiple different technology nodes from across each of the three major DRAM manufacturers.

TL;DR: Five state-of-the-art RowHammer mitigation mechanisms are evaluated using cycle-accurate simulation in the context of real data taken from chips to study how the mitigation mechanisms scale with chip vulnerability, and it is found that existing mechanisms either are not scalable or suffer from prohibitively large performance overheads in projected future devices.

TL;DR: BlockHammer as mentioned in this paper is a low-cost, effective, and easy-to-adopt RowHammer mitigation mechanism that prevents all RowHammers bit-flips while overcoming the two key challenges.