Loihi: A Neuromorphic Manycore Processor with On-Chip Learning

By Mike Davies, Narayan Srinivasa, Tsung-Han Lin, Gautham N. Chinya, Yongqiang Cao, S. H. Choday, G. Dimou, Prasad Joshi, Nabil Imam, Shweta Jain and 13 more. IEEE Micro, 2018

3,577 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/a67b7966cd8cdff0c9195cf719ac8ebf38c40b8e

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Electrical and Electronic Engineering

Citations in the corpus

Enabling Spike-Based Backpropagation for Training Deep Neural Network Architectures (2019, background): “Recently, specialized hardwares [38, 1, 8] have been developed to exploit this event-based, asynchronous signaling/processing scheme.”
Direct Training for Spiking Neural Networks: Faster, Larger, Better (2018, methodology): “To alleviate it, we adopt a voting strategy (Diehl and Matthew 2015) to decode the network output.”
Deep Learning With Spiking Neurons: Opportunities and Challenges (2018, methodology): “As an exception, we would like to mention the novel Intel Loihi chip (Davies et al., 2018), for which results of a single layer network on preprocessed”
Spike-driven Transformer (2023, methodology): “When implementing SNNs on neuromorphic chips, such as TrueNorth [2], Loihi [3], and Tianjic [4], only a small fraction of spiking neurons at any moment being active and the rest being idle.”
Long short-term memory and Learning-to-learn in networks of spiking neurons (2018, background): “The superior computing and learning capabilities of LSNNs suggest that they are also of interest for implementation in spike-based neuromorphic chips such as Brainscales [9], SpiNNaker [10], True North [2], chips from ETH Zürich [11], and Loihi [12].”
DIET-SNN: A Low-Latency Spiking Neural Network With Direct Input Encoding and Leakage and Threshold Optimization (2021, methodology): “The computations in SNN implemented on neuromorphic hardware are event-driven [2], [21].”
A 0.086-mm$^2$ 12.7-pJ/SOP 64k-Synapse 256-Neuron Online-Learning Digital Spiking Neuromorphic Processor in 28-nm CMOS (2018, methodology): “Author Schemmel [25] Benjamin [64] Qiao [23] Moradi [24] Park [65] Mayr [27] Painkras [16] Seo [28] Kim [29] Akopyan [30] Davies [31] Frenkel Publication ISCAS, 2010 PIEEE, 2014 Front.”
Hebbian Learning based Orthogonal Projection for Continual Learning of Spiking Neural Networks (2024, methodology): “…systems imitate biological neurons with parallel in-memory and event-driven neuronal computation and can be efficiently deployed on neuromorphic hardware for energy-efficient applications (Akopyan et al., 2015; Davies et al., 2018; Pei et al., 2019; Wo´zniak et al., 2020; Rao et al., 2022).”
Temporal Efficient Training of Spiking Neural Network via Gradient Re-weighting (2022, background): “The advantages of Spiking neuron networks (SNNs) lie in their energy-saving and fast-inference computation when embedded on neuromorphic hardware such as TrueNorth (DeBole et al., 2019) and Loihi (Davies et al., 2018).”
An artificial spiking afferent nerve based on Mott memristors for neurorobotics (2020, background): “Neuromorphic computing architecture based on spiking neural networks (SNNs) has been recognized as an attractive candidate for its promising energy ef ﬁ ciency and powerful computing capacity 6 – 8 .”
Synaptic Plasticity Dynamics for Deep Continuous Local Learning (DECOLLE) (2018, methodology): “Furthermore, forward-propagation techniques such as DECOLLE can be formulated as local synaptic plasticity rules, and are thus amenable to implementation in dedicated, event-based (neuromorphic) hardware [7] and compatible with neurobiology.”
Going Deeper With Directly-Trained Larger Spiking Neural Networks (2020, background): “ce (www.aaai.org). All rights reserved. consumption and better robustness(Stromatias et al. 2015). Moreover, SNNs can be easily applied on some specialized neuromorphic hardware (Merolla et al. 2014; Davies et al. 2018), which may be seen as the next generation of neural networks. There are two main”
Advancing Neuromorphic Computing With Loihi: A Survey of Results and Outlook (2021, methodology): “For more details on the Loihi chip architecture and implementation, we refer the readers to Davies et al. [13].”