3 papers
cs.AR2025
Towards Efficient LUT-based PIM: A Scalable and Low-Power Approach for Modern Workloads
Bahareh Khabbazan, Marc Riera, Antonio González
Data movement in memory-intensive workloads, such as deep learning, incurs energy costs that are over three orders of magnitude higher than the cost of computation. Since these wor…
cs.AR2025
Hamun: An Approximate Computation Method to Prolong the Lifespan of ReRAM-Based Accelerators
Mohammad Sabri, Marc Riera, Antonio Gonzalez
ReRAM-based accelerators exhibit enormous potential to increase computational efficiency for DNN inference tasks, delivering significant performance and energy savings over traditi…
cs.AR2024
ARAS: An Adaptive Low-Cost ReRAM-Based Accelerator for DNNs
Mohammad Sabri, Marc Riera, Antonio González
Processing Using Memory (PUM) accelerators have the potential to perform Deep Neural Network (DNN) inference by using arrays of memory cells as computation engines. Among various m…