Trustworthy tree-based machine learning by MoS2 flash-based analog content-addressable memory with inherent soft boundaries
flash-based analog CAM with inherent soft boundaries, enabling efficient inference with soft tree-based models. Our fabricated analog CAM arrays achieve 96% accuracy on the Wisconsin Diagnostic Breast Cancer dataset, and our experimentally calibrated model shows only a 0.6% accuracy drop on MNIST under 10% device threshold variation, compared with 45.3% for traditional decision trees.
Authors
- C W. LI
- Bo Wen (ORCID: https://orcid.org/0000-0002-6106-9509)
- Ruibin Mao (ORCID: https://orcid.org/0000-0001-6085-0486)
- Guoyun Gao (ORCID: https://orcid.org/0000-0003-2572-1413)
- Mingrui Jiang (ORCID: https://orcid.org/0009-0000-2364-3904)
- Xunzhao Yin (ORCID: https://orcid.org/0000-0003-4656-9545)
- Xiaobo Sharon Hu (ORCID: https://orcid.org/0000-0002-6636-9738)
- Xiaojuan Qi (ORCID: https://orcid.org/0000-0002-4285-1626)
- Zhicheng Xu (ORCID: https://orcid.org/0009-0001-2462-0162)
- Jiezhi Chen
Institutions
- University of Notre Dame (US)
- Hong Kong Baptist University (HK)
- Shandong University (CN)
- Education University of Hong Kong (HK)
- Zhejiang University (CN)
- University of Hong Kong (HK)
Publication Details
- Journal
- Nature Communications
- Published
- 2026-04-24
- DOI
- https://doi.org/10.1038/s41467-026-72118-z
- Primary Topic
- Network Packet Processing and Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Croucher Foundation
- Research Grants Council, University Grants Committee
- Innovation and Technology Commission