A Scalable Flexible SOM NoC-Based Hardware Architecture

Abadi, Mehdi; Jovanovic, Slavisa; Khalifa, Khaled Ben; Weber, S.; Bedoui, Mohamed Hédi

doi:10.1007/978-3-319-28518-4_14

Cited by 8 publications

(9 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lachmair et al ( [71]) propose a bus-interconnected multi-FPGA hardware SOM that allows the implementation of large reconfigurable networks. Abady et al propose a layered description of the HW SOM ( [72], [73]) where the computational neurons are decoupled from the communication that is implemented with a NoC. The same idea is exploited by Jovanovic et al ( [74], [75]) in a more hierarchical manner as the interconnected computational cells no longer implement neurons but clusters of neurons.…”

Section: Neuromorphic (Electronic) Architecturesmentioning

confidence: 99%

A unified software/hardware scalable architecture for brain-inspired computing based on self-organizing neural models

Muliukov,

Rodriguez,

Miramond

et al. 2022

Preprint

View full text Add to dashboard Cite

The field of artificial intelligence has significantly advanced over the past decades, inspired by discoveries from the fields of biology and neuroscience. The idea of this work is inspired by the process of self-organization of cortical areas in the human brain from both afferent and lateral/internal connections. In this work, we develop an original brain-inspired neural model associating Self-Organizing Maps (SOM) and Hebbian learning in the Reentrant SOM (ReSOM) model. The framework is applied to multimodal classification problems. Compared to existing methods based on unsupervised learning with post-labeling, the model enhances the state-of-the-art results. This work also demonstrates the distributed and scalable nature of the model through both simulation results and hardware execution on a dedicated FPGA-based platform named SCALP (Self-configurable 3D Cellular Adaptive Platform). SCALP boards can be interconnected in a modular way to support the structure of the neural model. Such a unified software and hardware approach enables the processing to be scaled and allows information from several modalities to be merged dynamically. The deployment on hardware boards provides performance results of parallel execution on several devices, with the communication between each board through dedicated serial links. The proposed unified architecture, composed of the ReSOM model and the SCALP hardware platform, demonstrates a significant increase in accuracy thanks to multimodal association, and a good trade-off between latency and power consumption compared to a centralized GPU implementation.

show abstract

Section: Neuromorphic (Electronic) Architecturesmentioning

confidence: 99%

A unified software/hardware scalable architecture for brain-inspired computing based on self-organizing neural models

Muliukov,

Rodriguez,

Miramond

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Lachmair et al ( 2013 ) propose a bus-interconnected multi-FPGA hardware SOM that allows the implementation of large reconfigurable networks. Abady et al propose a layered description of the HW SOM (Abadi et al, 2016 , 2018 ) where the computational neurons are decoupled from the communication that is implemented with a NoC. The same idea is exploited in Jovanović et al ( 2018 ) and Jovanović et al ( 2020 ) in a more hierarchical manner as the interconnected computational cells no longer implement neurons but clusters of neurons.…”

Section: State Of the Artmentioning

confidence: 99%

A Unified Software/Hardware Scalable Architecture for Brain-Inspired Computing Based on Self-Organizing Neural Models

et al. 2022

View full text Add to dashboard Cite

The field of artificial intelligence has significantly advanced over the past decades, inspired by discoveries from the fields of biology and neuroscience. The idea of this work is inspired by the process of self-organization of cortical areas in the human brain from both afferent and lateral/internal connections. In this work, we develop a brain-inspired neural model associating Self-Organizing Maps (SOM) and Hebbian learning in the Reentrant SOM (ReSOM) model. The framework is applied to multimodal classification problems. Compared to existing methods based on unsupervised learning with post-labeling, the model enhances the state-of-the-art results. This work also demonstrates the distributed and scalable nature of the model through both simulation results and hardware execution on a dedicated FPGA-based platform named SCALP (Self-configurable 3D Cellular Adaptive Platform). SCALP boards can be interconnected in a modular way to support the structure of the neural model. Such a unified software and hardware approach enables the processing to be scaled and allows information from several modalities to be merged dynamically. The deployment on hardware boards provides performance results of parallel execution on several devices, with the communication between each board through dedicated serial links. The proposed unified architecture, composed of the ReSOM model and the SCALP hardware platform, demonstrates a significant increase in accuracy thanks to multimodal association, and a good trade-off between latency and power consumption compared to a centralized GPU implementation.

show abstract

“…Our results are compared to three FPGA SOM implementations which are of particular interest in our case because they are all based on a NoC-based FPGA implementations [10,11,23].…”

Section: Performance Comparisonmentioning

confidence: 99%

“…The use of Euclidean distance used in the formal definition of SOM and in most of its software implementations are often replaced by a Manhattan distance [6][7][8][9][10] or Chessboard [6] distance functions to avoid multipliers and square root operators. Another common optimization is to replace the exponentiation used to compute the Gaussian neighboring function by approximations such as negative power of two implemented with shift registers [5,6,10,11] or Lookup-Tables [7,9]. In all the previously cited works, vectors are represented with integer or fixed-point arithmetic, with usually 16-bits per vector component.…”

Section: Introductionmentioning

confidence: 99%

Hardware Architecture for Asynchronous Cellular Self-Organizing Maps

2022

View full text Add to dashboard Cite

Nowadays, one of the main challenges in computer architectures is scalability; indeed, novel processor architectures can include thousands of processing elements on a single chip and using them efficiently remains a big issue. An interesting source of inspiration for handling scalability is the mammalian brain and different works on neuromorphic computation have attempted to address this question. The Self-configurable 3D Cellular Adaptive Platform (SCALP) has been designed with the goal of prototyping such types of systems and has led to the proposal of the Cellular Self-Organizing Maps (CSOM) algorithm. In this paper, we present a hardware architecture for CSOM in the form of interconnected neural units with the specific property of supporting an asynchronous deployment on a multi-FPGA 3D array. The Asynchronous CSOM (ACSOM) algorithm exploits the underlying Network-on-Chip structure to be provided by SCALP in order to overcome the multi-path propagation issue presented by a straightforward CSOM implementation. We explore its behaviour under different map topologies and scalar representations. The results suggest that a larger network size with low precision coding obtains an optimal ratio between algorithm accuracy and FPGA resources.

show abstract

A Scalable Flexible SOM NoC-Based Hardware Architecture

Cited by 8 publications

References 13 publications

A unified software/hardware scalable architecture for brain-inspired computing based on self-organizing neural models

A unified software/hardware scalable architecture for brain-inspired computing based on self-organizing neural models

A Unified Software/Hardware Scalable Architecture for Brain-Inspired Computing Based on Self-Organizing Neural Models

Hardware Architecture for Asynchronous Cellular Self-Organizing Maps

Contact Info

Product

Resources

About