DIgital-ANAlog (DIANA), a heterogeneous multi-core accelerator, combines a reduced instruction set computer-five (RISC-V) host processor with an analog in-memory computing …
The rapidly-changing deep learning landscape presents a unique opportunity for building inference accelerators optimized for specific datacenter-scale workloads. We propose Full …
In the hardware design space exploration process, it is critical to optimize both hardware parameters and algorithm-to-hardware mappings. Previous work has largely approached …
DNN workloads can be scheduled onto DNN accelerators in many different ways: from layer- by-layer scheduling to cross-layer depth-first scheduling (aka layer fusion, or cascaded …
V Jain, S Giraldo, J De Roose, L Mei… - IEEE Journal of Solid …, 2023 - ieeexplore.ieee.org
Extreme edge devices or Internet-of-Things (IoT) nodes require both ultra-low power (ULP) always-on (AON) processing as well as the ability to do on-demand sampling and …
Memory buffer allocation for on-chip memories is a major challenge in modern machine learning systems that target ML accelerators. In interactive systems such as mobile phones …
C Sakhuja, Z Shi, C Lin - 2023 IEEE International Symposium …, 2023 - ieeexplore.ieee.org
Deep learning accelerators are important tools for feeding the growing demand for deep learning applications. The automated design of such accelerators—which is important for …
In-memory-computing is emerging as an efficient hardware paradigm for deep neural network accelerators at the edge, enabling to break the memory wall and exploit massive …
Map Space Exploration is the problem of finding optimized mappings of a Deep Neural Network (DNN) model on an accelerator. It is known to be extremely computationally …