LionHeart: A Layer-based Mapping Framework for Heterogeneous Systems with Analog In-Memory Computing Tiles
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908278362996736 |
|---|---|
| author | Lammie, Corey Wang, Yuxuan Ponzina, Flavio Klein, Joshua Benmeziane, Hadjer Zapater, Marina Boybat, Irem Sebastian, Abu Ansaloni, Giovanni Atienza, David |
| author_facet | Lammie, Corey Wang, Yuxuan Ponzina, Flavio Klein, Joshua Benmeziane, Hadjer Zapater, Marina Boybat, Irem Sebastian, Abu Ansaloni, Giovanni Atienza, David |
| contents | When arranged in a crossbar configuration, resistive memory devices can be used to execute Matrix-Vector Multiplications (MVMs), the most dominant operation of many Machine Learning (ML) algorithms, in constant time complexity. Nonetheless, when performing computations in the analog domain, novel challenges are introduced in terms of arithmetic precision and stochasticity, due to non-ideal circuit and device behaviour. Moreover, these non-idealities have a temporal dimension, resulting in a degrading application accuracy over time. Facing these challenges, we propose a novel framework, named LionHeart, to obtain hybrid analog-digital mappings to execute Deep Learning (DL) inference workloads using heterogeneous accelerators. The accuracy-constrained mappings derived by LionHeart showcase, across different Convolutional Neural Networks (CNNs) and one transformer-based network, high accuracy and potential for speedup. The results of the full system simulations highlight run-time reductions and energy efficiency gains that exceed 6X, with a user-defined accuracy threshold for a fully digital floating point implementation. LionHeart is open-sourced here: https://github.com/IBM/lionheart. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_09420 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | LionHeart: A Layer-based Mapping Framework for Heterogeneous Systems with Analog In-Memory Computing Tiles Lammie, Corey Wang, Yuxuan Ponzina, Flavio Klein, Joshua Benmeziane, Hadjer Zapater, Marina Boybat, Irem Sebastian, Abu Ansaloni, Giovanni Atienza, David Emerging Technologies When arranged in a crossbar configuration, resistive memory devices can be used to execute Matrix-Vector Multiplications (MVMs), the most dominant operation of many Machine Learning (ML) algorithms, in constant time complexity. Nonetheless, when performing computations in the analog domain, novel challenges are introduced in terms of arithmetic precision and stochasticity, due to non-ideal circuit and device behaviour. Moreover, these non-idealities have a temporal dimension, resulting in a degrading application accuracy over time. Facing these challenges, we propose a novel framework, named LionHeart, to obtain hybrid analog-digital mappings to execute Deep Learning (DL) inference workloads using heterogeneous accelerators. The accuracy-constrained mappings derived by LionHeart showcase, across different Convolutional Neural Networks (CNNs) and one transformer-based network, high accuracy and potential for speedup. The results of the full system simulations highlight run-time reductions and energy efficiency gains that exceed 6X, with a user-defined accuracy threshold for a fully digital floating point implementation. LionHeart is open-sourced here: https://github.com/IBM/lionheart. |
| title | LionHeart: A Layer-based Mapping Framework for Heterogeneous Systems with Analog In-Memory Computing Tiles |
| topic | Emerging Technologies |
| url | https://arxiv.org/abs/2401.09420 |