This paper presents a novel technique to reduce energy consumption of a machine learning classifier based on incremental-precision feature computation and classification. Specifically, the algorithm starts with features computed using the lowest possible precision. Depending on the classification accuracy, the features of the previous level are combined with features of the incremental-precision to compute the features in higher-precision. This process is continued till a desired accuracy is obtained. A certain threshold that allows many samples to be classified using a low-precision classifier can reduce energy consumption, but increases misclassification error. To implement hardware which provides the required updates in precision, an incremental-precision architecture based on data-path decomposition is proposed. One novel aspect of this work lies in the design of appropriate thresholds for multi-level classification using training data such that a family of designs can be obtained that enable trade-offs between classification accuracy and energy consumption. Another novel aspect involves the design of hardware architectures based on data-path decomposition which can incrementally increase precision upon demand. Using a seizure detection example, it is shown that the proposed incremental-precision-based multi-level classification approach can reduce energy consumption by 35% while maintaining high sensitivity, or by about 50% at the expense of 15% degradation in sensitivity compared with similar approaches to seizure detection in literature. The reduction in energy is achieved at the expense of small area, timing and memory overheads as multiple classification steps are used instead of a single step.
|Original language||English (US)|
|Number of pages||14|
|Journal||IEEE Journal on Emerging and Selected Topics in Circuits and Systems|
|State||Published - Dec 2018|
Bibliographical noteFunding Information:
Manuscript received January 3, 2018; revised April 2, 2018; accepted April 25, 2018. Date of publication May 14, 2018; date of current version December 11, 2018. This work was supported by the National Science Foundation under Grant CCF-1749494. This paper was recommended by Guest Editor M. Alioto. (Corresponding author: Keshab K. Parhi.) The authors are with the University of Minnesota Twin Cities, Minneapolis, MN 55455 USA (e-mail: firstname.lastname@example.org).
- Approximate computing
- Datapath decomposition
- Energy reduction
- Incremental-precision computation
- Machine learning
- Multi-level classification
- Seizure detection