IEEE Micro· 2018Q1
Loihi: A Neuromorphic Manycore Processor with On-Chip Learning
- 3,762citations
- Q1SCImago
- 2018year
Short summary
Intel's Loihi chip, a 60-mm2 neuromorphic processor, achieves over 1000x better energy-delay product than CPUs for LASSO optimization using spiking neural networks with on-chip learning.
AI-generated from the title and abstract; the full text is not read.
Key points
- Loihi is a 60-mm2 neuromorphic processor with advanced SNN modeling features like dendritic compartments and synaptic delays.
- It incorporates programmable synaptic learning rules, enabling on-chip learning.
- Loihi solves LASSO optimization problems using a spiking convolutional algorithm.
- It achieves over 1000x better energy-delay product compared to conventional CPU solvers for LASSO optimization.
AI-generated from the title and abstract; the full text is not read.
Abstract
Loihi is a 60-mm2 chip fabricated in Intels 14-nm process that advances the state-of-the-art modeling of spiking neural networks in silicon. It integrates a wide range of novel features for the field, such as hierarchical connectivity, dendritic compartments, synaptic delays, and, most importantly, programmable synaptic learning rules. Running a spiking convolutional form of the Locally Competitive Algorithm, Loihi can solve LASSO optimization problems with over three orders of magnitude superior energy-delay-product compared to conventional solvers running on a CPU iso-process/voltage/area. This provides an unambiguous example of spike-based computation, outperforming all known conventional solutions.
The authors' abstract, as published at the source. IEEE Micro, 2018 · DOI ↗
Continue with a free account
Ask the paper: 3 free questions a day about this paper; save it, get its citation, new summaries every day for your field. Takeaways are Premium.
Continue free on the webSign in with Google or Apple; no card needed. You come back to this paper.
On your phone:
Field: Electrical and Electronic Engineering
Electrical and Electronic EngineeringEngineering