Summary of A Tiny Supervised Odl Core with Auto Data Pruning For Human Activity Recognition, by Hiroki Matsutani et al.
A Tiny Supervised ODL Core with Auto Data Pruning for Human Activity Recognition
by Hiroki Matsutani, Radu Marculescu
First submitted to arxiv on: 2 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Hardware Architecture (cs.AR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces a low-cost and low-power tiny supervised on-device learning (ODL) core for human activity recognition, which can address the distributional shift of input data. The ODL core combines automatic data pruning with supervised ODL to reduce queries needed to acquire predicted labels from a nearby teacher device, saving power consumption during model retraining. The proposed solution uses a 45nm CMOS process technology and requires less memory size than a multilayer perceptron (MLP). Experiments show that the automatic data pruning reduces communication volume by 55.7% with only 0.9% accuracy loss. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper develops a tiny machine learning device for recognizing human activities, which can be powered by just a few milliwatts. The device uses a new way to learn and adapt to changing data without needing a lot of power or memory. This makes it useful for devices that need to recognize and respond to human actions in real-time, such as smart home systems or wearable fitness trackers. |
Keywords
» Artificial intelligence » Activity recognition » Machine learning » Pruning » Supervised
