Cellular Self-Organising Maps - CSOM

Girau, Bernard; Upegui, Andrés

doi:10.1007/978-3-030-19642-4_4

Cited by 4 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to the direct exchange of information between neighboring SOM neurons, we can reduce the algorithm's complexity from O ( n ) to

, where n is the number of SOM neurons. In future, it is planned to correct this drawback and integrate one of the previously proposed Iterative Grid (Rodriguez et al, 2018 ) or Cellular SOM (Girau and Upegui, 2019 ) solutions in order to significantly speed up the calculations.…”

Section: Discussionmentioning

confidence: 99%

“…So it becomes possible to dynamically adapt the size of the network to the data structure, but at the cost of a more expensive learning process. The C(ellular) SOM model (CSOM) (Girau and Upegui, 2019 ) also offers gains in accuracy and energy consumption, using a simplified grid to enhance its hardware implementation. The Pruning Cellular Self-Organizing Maps (PCSOM) model (Upegui et al, 2018 ) prunes some SOM connections to gain better performance.…”

Section: State Of the Artmentioning

confidence: 99%

See 1 more Smart Citation

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

et al. 2022

Self Cite

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

“…Due to the direct exchange of information between neighboring SOM neurons, we can reduce the algorithm's complexity from O ( n ) to

Section: Discussionmentioning

confidence: 99%

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

Self Cite

View full text Add to dashboard Cite

show abstract

“…So it becomes possible to dynamically adapt the size of the network to the data structure, but at the cost of a more expensive learning process. The C(ellular) SOM model (CSOM) [37] also offers gains in accuracy and energy consumption, using a simplified grid to enhance its hardware implementation. The Pruning Cellular Self-Organizing Maps (PCSOM) model [38] prunes some SOM connections to gain better performance.…”

Section: A Brain-inspired Self-organizing Neural Modelsmentioning

confidence: 99%

“…1) Hardware development: A distinctive feature of SCALP is an integrated FPGA chip, which enables a grid of independent computing elements to be built. This paper ignores this feature of the boards, but some solutions have previously been proposed [20], [37]. They develop the idea to use the FPGA for executing a SOM and accelerating the algorithm for calculating its BMU.…”

Section: B Possible Development Directionsmentioning

confidence: 99%

See 1 more Smart Citation

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

Muliukov,

Rodriguez,

Miramond

et al. 2022

Preprint

Self Cite

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

Statistical Perspective of SOM and CSOM for Hyper-Spectral Image Classification

Mallapragada

Hung

2020

IGARSS 2020 - 2020 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

Cellular Self-Organising Maps - CSOM

Cited by 4 publications

References 9 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

Statistical Perspective of SOM and CSOM for Hyper-Spectral Image Classification

Contact Info

Product

Resources

About