MonoSparse-CAM: Efficient Tree Model Processing via Monotonicity and Sparsity in CAMs
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913626000982016 |
|---|---|
| author | Molom-Ochir, Tergel Taylor, Brady Li, Hai Chen, Yiran |
| author_facet | Molom-Ochir, Tergel Taylor, Brady Li, Hai Chen, Yiran |
| contents | While the tree-based machine learning (TBML) models exhibit superior performance compared to neural networks on tabular data and hold promise for energy-efficient acceleration using aCAM arrays, their ideal deployment on hardware with explicit exploitation of TBML structure and aCAM circuitry remains a challenging task. In this work, we present MonoSparse-CAM, a new CAM-based optimization technique that exploits TBML sparsity and monotonicity in CAM circuitry to further advance processing performance. Our results indicate that MonoSparse-CAM reduces energy consumption by upto to 28.56x compared to raw processing and by 18.51x compared to state-of-the-art techniques, while improving the efficiency of computation by at least 1.68x. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_11071 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MonoSparse-CAM: Efficient Tree Model Processing via Monotonicity and Sparsity in CAMs Molom-Ochir, Tergel Taylor, Brady Li, Hai Chen, Yiran Machine Learning Artificial Intelligence Hardware Architecture While the tree-based machine learning (TBML) models exhibit superior performance compared to neural networks on tabular data and hold promise for energy-efficient acceleration using aCAM arrays, their ideal deployment on hardware with explicit exploitation of TBML structure and aCAM circuitry remains a challenging task. In this work, we present MonoSparse-CAM, a new CAM-based optimization technique that exploits TBML sparsity and monotonicity in CAM circuitry to further advance processing performance. Our results indicate that MonoSparse-CAM reduces energy consumption by upto to 28.56x compared to raw processing and by 18.51x compared to state-of-the-art techniques, while improving the efficiency of computation by at least 1.68x. |
| title | MonoSparse-CAM: Efficient Tree Model Processing via Monotonicity and Sparsity in CAMs |
| topic | Machine Learning Artificial Intelligence Hardware Architecture |
| url | https://arxiv.org/abs/2407.11071 |