GPU-Based Self-Organizing Maps for Post-labeled Few-Shot Unsupervised Learning

Khacef, Lyes; Gripon, Vincent; Miramond, Benoît

doi:10.1007/978-3-030-63833-7_34

Cited by 7 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Notice that our model has used the datasets in a raw form, without any feature extractor, which might radically enhance the quality of the vectors representing the data. The addition of feature extractors as input of the neural maps increases the algorithm's accuracy, as we showed in [32], [76].…”

Section: A Simulation Resultsmentioning

confidence: 77%

“…In parallel, it has been shown that SOMs perform better at representing overlapping structures compared to classical clustering techniques such as partitive clustering or K-means [31]. Recently, we showed that the SOM combined with transfer learning reaches a competitive accuracy on complex few-shot learning problems [32]. In addition, SOMs were directly inspired by work on the cerebral cortex [4], and can hence be potentially applied in the future for the development of biologically compatible components.…”

Section: A Brain-inspired Self-organizing Neural Modelsmentioning

confidence: 99%

“…Lastly, it should be noted that the problem addressed by our framework does not correspond to a typical SSL problem, since the existing labels are available (or provided) only after the training. So we consider a special type of SSL problem, called in the literature the Post-Labeled Unsupervised Learning problem [32], [76].…”

Section: Application Of Resom To Classification Tasksmentioning

confidence: 99%

See 2 more Smart Citations

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

Section: A Simulation Resultsmentioning

confidence: 77%

Section: A Brain-inspired Self-organizing Neural Modelsmentioning

confidence: 99%

Section: Application Of Resom To Classification Tasksmentioning

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

show abstract

“…In parallel, it has been shown that SOMs perform better at representing overlapping structures compared to classical clustering techniques such as partitive clustering or K-means (Budayan et al, 2009 ). Recently, we showed that the SOM combined with transfer learning reaches a competitive accuracy on complex few-shot learning problems (Khacef et al, 2020a ). In addition, SOMs were directly inspired by work on the cerebral cortex (Kohonen, 1990 ), and can hence be potentially applied in the future for the development of biologically compatible components.…”

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

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

“…The overall design of the cyberinfrastructure and the SOM-based data analytics is illustrated in Figure 1. For the cyberinfrastructure implementation, we propose the adoption of the Tensorflow-based SOM package (Khacef et al, 2020) in this project, as the package can be readily deployed on the Summit supercomputer at ORNL.…”

Section: Narrativementioning

confidence: 99%

A HPC Theory-Guided Machine Learning Cyberinfrastructure for Communicating Hydrometeorological Data Across Scales

Allen‐Dumas

Berres

et al. 2021

View full text Add to dashboard Cite

Focal Area(s)(2) Predictive modeling through the use of AI techniques and AI-derived model components. In this white paper, we propose the design and development of a high-performance computing (HPC) -powered cyberinfrastructure to provide a computation-efficient downscaling solution that can interpolate high-resolution predictions of hydrometeorological variables (e.g., precipitation and flood depth grid) using low-resolution raster-based simulation outputs. The cyberinfrastructure adopts a theory-guided machine learning approach to enable timely and high-resolution hydrometeorological predictions for supporting decisions in hydropower operations and mitigation of multiple water-related hazards (e.g., flood, sedimentation, stream bank instability).

show abstract

GPU-Based Self-Organizing Maps for Post-labeled Few-Shot Unsupervised Learning

Cited by 7 publications

References 19 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

A HPC Theory-Guided Machine Learning Cyberinfrastructure for Communicating Hydrometeorological Data Across Scales

Contact Info

Product

Resources

About