Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering
TinyML enables the deployment of Machine Learning (ML) models on resource-constrained devices, addressing a growing need for efficient, low-power AI solutions. However, significant challenges remain due to strict memory, processing, and energy limitations. This study introduces a novel method to opt...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10849357/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832576734485020672 |
|---|---|
| author | Thommas K. S. Flores Morsinaldo Medeiros Marianne Silva Daniel G. Costa Ivanovitch Silva |
| author_facet | Thommas K. S. Flores Morsinaldo Medeiros Marianne Silva Daniel G. Costa Ivanovitch Silva |
| author_sort | Thommas K. S. Flores |
| collection | DOAJ |
| description | TinyML enables the deployment of Machine Learning (ML) models on resource-constrained devices, addressing a growing need for efficient, low-power AI solutions. However, significant challenges remain due to strict memory, processing, and energy limitations. This study introduces a novel method to optimize Post-Training Quantization (PTQ), a widely used technique for reducing model size, by integrating Vector Quantization (VQ) with incremental clustering. While VQ is a technique that reduces model size by grouping similar parameters, incremental clustering, implemented via the AutoCloud K-Fixed algorithm, preserves accuracy during compression. This combined approach was validated on an automotive dataset predicting CO2 emissions from vehicle sensor measurements such as mass air flow, intake pressure, temperature, and speed. The model was quantized and deployed on Macchina A0 hardware, demonstrating over 90% compression with negligible accuracy loss. Results show improved performance and deployment efficiency, showcasing the potential of this combined technique for real-world embedded applications. |
| format | Article |
| id | doaj-art-0791b23d8ce6423699c994247ff5e615 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-0791b23d8ce6423699c994247ff5e6152025-01-31T00:01:59ZengIEEEIEEE Access2169-35362025-01-0113174401745610.1109/ACCESS.2025.353284910849357Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental ClusteringThommas K. S. Flores0https://orcid.org/0000-0003-2808-8529Morsinaldo Medeiros1https://orcid.org/0000-0001-7624-5301Marianne Silva2https://orcid.org/0000-0002-8277-7571Daniel G. Costa3https://orcid.org/0000-0003-3988-8476Ivanovitch Silva4https://orcid.org/0000-0002-0116-6489PPgEEC, Federal University of Rio Grande do Norte, Natal, BrazilPPgEEC, Federal University of Rio Grande do Norte, Natal, BrazilSI, Federal University of Alagoas, Penedo, BrazilSYSTEC-ARISE, Faculty of Engineering, University of Porto, Porto, PortugalPPgEEC, Federal University of Rio Grande do Norte, Natal, BrazilTinyML enables the deployment of Machine Learning (ML) models on resource-constrained devices, addressing a growing need for efficient, low-power AI solutions. However, significant challenges remain due to strict memory, processing, and energy limitations. This study introduces a novel method to optimize Post-Training Quantization (PTQ), a widely used technique for reducing model size, by integrating Vector Quantization (VQ) with incremental clustering. While VQ is a technique that reduces model size by grouping similar parameters, incremental clustering, implemented via the AutoCloud K-Fixed algorithm, preserves accuracy during compression. This combined approach was validated on an automotive dataset predicting CO2 emissions from vehicle sensor measurements such as mass air flow, intake pressure, temperature, and speed. The model was quantized and deployed on Macchina A0 hardware, demonstrating over 90% compression with negligible accuracy loss. Results show improved performance and deployment efficiency, showcasing the potential of this combined technique for real-world embedded applications.https://ieeexplore.ieee.org/document/10849357/Embedded systemspost-training quantizationvector quantizationautomotive sensorspollution |
| spellingShingle | Thommas K. S. Flores Morsinaldo Medeiros Marianne Silva Daniel G. Costa Ivanovitch Silva Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering IEEE Access Embedded systems post-training quantization vector quantization automotive sensors pollution |
| title | Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering |
| title_full | Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering |
| title_fullStr | Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering |
| title_full_unstemmed | Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering |
| title_short | Enhanced Vector Quantization for Embedded Machine Learning: A Post-Training Approach With Incremental Clustering |
| title_sort | enhanced vector quantization for embedded machine learning a post training approach with incremental clustering |
| topic | Embedded systems post-training quantization vector quantization automotive sensors pollution |
| url | https://ieeexplore.ieee.org/document/10849357/ |
| work_keys_str_mv | AT thommasksflores enhancedvectorquantizationforembeddedmachinelearningaposttrainingapproachwithincrementalclustering AT morsinaldomedeiros enhancedvectorquantizationforembeddedmachinelearningaposttrainingapproachwithincrementalclustering AT mariannesilva enhancedvectorquantizationforembeddedmachinelearningaposttrainingapproachwithincrementalclustering AT danielgcosta enhancedvectorquantizationforembeddedmachinelearningaposttrainingapproachwithincrementalclustering AT ivanovitchsilva enhancedvectorquantizationforembeddedmachinelearningaposttrainingapproachwithincrementalclustering |