Smart Sensor Calibration with Auto-Rotating Perceptrons
Sensor calibration is vital to have valid measure-ments of physical activities. Herein we deal withadjusting the signal from a wearable force sensoragainst a reference scale. By using a few samplesand data augmentation, we trained a neural-basedregression model to correct the wearable output.For this task, we tested the novel Auto-RotatingPerceptrons (ARP). We found that a neural ARPmodel with sigmoid activations can outperforman identical neural network based on classic per-ceptrons with sigmoid and even ReLU activation.
Sensor calibration is vital to have valid measure-ments of physical activities. Herein we deal withadjusting the signal from a wearable force sensoragainst a reference scale. By using a few samplesand data augmentation, we trained a neural-basedregression model to correct the wearable output.For this task, we tested the novel Auto-RotatingPerceptrons (ARP). We found that a neural ARPmodel with sigmoid activations can outperforman identical neural network based on classic per-ceptrons with sigmoid and even ReLU activation.
Año: 2020
Título de la revista: LatinX in AI Research Workshop at ICML 2020.
Url: https://www.researchgate.net/publication/343413813_Smart_Sensor_Calibration_with_Auto-Rotating_Perceptrons