Summary of Domino: Domain-invariant Hyperdimensional Classification For Multi-sensor Time Series Data, by Junyao Wang et al.
DOMINO: Domain-invariant Hyperdimensional Classification for Multi-Sensor Time Series Data
by Junyao Wang, Luke Chen, Mohammad Abdullah Al Faruque
First submitted to arxiv on: 7 Aug 2023
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
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 This paper proposes a novel machine learning framework called DOMINO for edge-based learning on noisy multi-sensor time-series data. DOMINO addresses the distribution shift problem by leveraging efficient and parallel matrix operations in high-dimensional space to identify and filter out domain-variant dimensions. The proposed approach achieves state-of-the-art accuracy, with an average 2.04% improvement over deep neural network (DNN) based domain generalization techniques. Additionally, DOMINO delivers faster training and inference speeds, at 16.34x and 2.89x respectively, making it a more efficient solution for edge devices. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary DOMINO is a new way to do machine learning on tiny computers that collect data from many sensors. This type of computer is called an “edge device”. When we train a model on data, we want it to work well even if the data changes in some way. This problem is called “distribution shift”. DOMINO solves this problem by using special math operations to filter out the parts of the data that are different from what the model was trained on. It does this really fast and accurately, beating other models at doing this task. |
Keywords
* Artificial intelligence * Domain generalization * Inference * Machine learning * Neural network * Time series
