Benchmark Comparisons of Spike-based Reconfigurable Neuroprocessor Architectures for Control Applications

Foshie, Adam Z.; Rizzo, Charles; Das, Hritom; Zheng, Chaohui; Plank, James S.; Rose, Garrett S.

doi:10.1145/3526241.3530381

Cited by 9 publications

(5 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We employed RAVENS for our neuromorphic hardware implementation [24,39]. RAVENS is a neuroprocessor for SNNs which has implementations in CPU Simulation, FPGA, and micro-controllers.…”

Section: Hardware Implementationmentioning

confidence: 99%

“…RAVENS includes features such as I&F neuron model, linear charge leakage, axonal delay, adjustable resting potential, and programmable absolute and relative refractory period. Readers may refer to [39] for a detailed overview of the RAVENS neuroprocessor. The feature of RAVENS we target in this work is its STDP implementation.…”

Section: Hardware Implementationmentioning

confidence: 99%

“…SNNs tailored for specific applications are constructed through a blend of supervised learning, employing genetic algorithms such as evolutionary optimization (EO), and unsupervised learning through STDP. The physical implementation of the SNN is achieved using RAVENS [24].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Hardware software co-design for leveraging STDP in a memristive neuroprocessor

Chakraborty,

Ameli,

Das

et al. 2024

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

In neuromorphic computing, different learning mechanisms are being widely adopted to improve the performance of a specific application. Among these techniques, Spike-Timing-Dependent Plasticity (STDP) stands out as one of the most favored. STDP is simply managed by the temporal information of an event, which is biologically inspired. However, most of the prior works on STDP are focused on circuit implementation or software simulation for performance evaluation. Previous works also lack a comparative analysis of the performances of different STDP implementations. This study aims to provide a comprehensive assessment of STDP, centering on the performance across various applications such as classification (static and temporal datasets), control, and reservoir computing. Different applications necessitate distinct STDP configurations to achieve optimal performance with the neuroprocessor. Additionally, this work introduces an Application-Specific Integrated Circuit (ASIC) design of STDP circuitry. The design is based on current-controlled memristive synapse principles and utilizes 65nm CMOS technology from IBM. The detailed presentation includes circuitry specifics, layout, and performance parameters such as energy consumption and design area.

show abstract

“…We employed RAVENS for our neuromorphic hardware implementation [24,39]. RAVENS is a neuroprocessor for SNNs which has implementations in CPU Simulation, FPGA, and micro-controllers.…”

Section: Hardware Implementationmentioning

confidence: 99%

Section: Hardware Implementationmentioning

confidence: 99%

See 1 more Smart Citation

Hardware software co-design for leveraging STDP in a memristive neuroprocessor

Chakraborty,

Ameli,

Das

et al. 2024

Neuromorph. Comput. Eng.

View full text Add to dashboard Cite

show abstract

“…In the domain of (neuro)robotics [79], viable tools have emerged over the past years [115,366,226], offering improved physics engines for simulating real-world robotic systems and some of the task environments in which they would operate. With respect to neuromorphic substrates, simulators of neuromorphic chips [196,217] and emulator kits, e.g., the RISP neuro-processor [282,281,130], have recently emerged; offering an alternative to expensive, non-mainstream hardware. However, while digital emulation might prove to be an invaluable solution, even serving as a force behind the rapid prototyping of early mortal computers, there is always a prescient gap between the fidelity and realism of a particular simulation and the real-world niche that it attempts to emulate, i.e., the "sim2real" problem [173].…”

Section: A3 On Neuromorphic Systems and The Body-niche Problemmentioning

confidence: 99%

Mortal Computation: A Foundation for Biomimetic Intelligence

Ororbia,

Friston

2023

Preprint

View full text Add to dashboard Cite

This review motivates and synthesizes research efforts in neuroscience-inspired artificial intelligence and biomimetic computing in terms of mortal computation. Specifically, we characterize the notion of mortality by recasting ideas in biophysics, cybernetics, and cognitive science in terms of a theoretical foundation for sentient behavior. We frame the mortal computation thesis through the Markov blanket formalism and the circular causality entailed by inference, learning, and selection. The ensuing framework -- underwritten by the free energy principle -- could prove useful for guiding the construction of unconventional connectionist computational systems, neuromorphic intelligence, and chimeric agents, including sentient organoids, which stand to revolutionize the long-term future of embodied, enactive artificial intelligence and cognition research.

show abstract

“…In our prototype implementation, we have written simulators of two neuroprocessors developed by the TENNLab research team: RAVENS [7] and RISP [17]. Both employ clocked integrate-and-fire neurons, and synapses with integer delays.…”

Section: B the Neuroprocessormentioning

confidence: 99%