A survey on memory subsystems for deep neural network accelerators

A Asad, R Kaur, F Mohammadi - Future Internet, 2022 - mdpi.com
From self-driving cars to detecting cancer, the applications of modern artificial intelligence
(AI) rely primarily on deep neural networks (DNNs). Given raw sensory data, DNNs are able …

Processing Multi-Layer Perceptrons In-Memory

P Carrinho, O Ferraz, JD Ferreira… - … IEEE Workshop on …, 2024 - ieeexplore.ieee.org
Important modern applications such as machine learning, deep learning, graph processing,
databases (and many others) are memory-bound. This creates a bottleneck caused by the …

Custom Memory Design for Logic-in-Memory: Drawbacks and Improvements over Conventional Memories

F Ottati, G Turvani, G Masera, M Vacca - Electronics, 2021 - mdpi.com
The speed of modern digital systems is severely limited by memory latency (the “Memory
Wall” problem). Data exchange between Logic and Memory is also responsible for a large …

Heterogeneous Uncore Architectures in Future Chip-Multi Processors

A Asad - rshare.library.torontomu.ca
Uncore components including on-chip memory systems and interconnects consume a
significant portion of overall energy consumption in emerging embedded applications …

[PDF][PDF] Custom Memory Design for Logic-in-Memory: Drawbacks and Improvements over Conventional Memories. Electronics 2021, 10, 2291

F Ottati, G Turvani, G Masera, M Vacca - 2021 - academia.edu
The speed of modern digital systems is severely limited by memory latency (the “Memory
Wall” problem). Data exchange between Logic and Memory is also responsible for a large …