Mobile Edge Computing (MEC) powered by Unmanned Aerial Vehicles (UAVs) can provide ubiquitous communication and computing services for the Internet of Things (IoT). Integrating Intelligent Reflecting Surface (IRS) wit...
详细信息
Minimum miscibility pressure (MMP) prediction plays an important role in design and operation of nitrogen based enhanced oil recovery processes. In this work, a comparative study of statistical and machine learning me...
详细信息
Mechanical properties are one of the most critical indicators of hot-rolled seamless steel pipe. Seamless steel pipe served in harsh working environments for a long time, putting strict requirements on product quality...
详细信息
Differential drive mobile robots (DDMRs) are often used in automation research because they are simple mechanically, and low cost as they only require two motors to drive and steer. Work to date has focused on control...
详细信息
At present, many domestic scholars and research institutions have studied the laser vision welding seam tracking system. By improving the control strategy of the robot and welding seam tracking system, the processing ...
详细信息
This topic studies the gait control and attitude control of a quadruped robot during flat slope movement. It studies its motion control strategies and methods from the aspects of model construction, gait analysis, and...
详细信息
At present, biomass fuel is often injected into the rotary kiln through the ignition channel or coal fuel channel of traditional burners for cement production. However, neither of these channels is designed for the in...
详细信息
Parkinson's disease is a relatively common neurodegenerative disorder that causes motor impairments, with gait disturbances being a prominent symptom. Gait analysis helps physicians accurately assess the condition...
详细信息
This study investigates the problem of distributed adaptive formation control of connected vehicles with actuator saturation and time-varying spacing. Firstly, optimization performance metrics are defined based on the...
详细信息
Artificial neural network (ANN) has been widely used in automation. However, the vulnerability of ANN under certain attacks poses security threat for critical automation systems. Adding noises to artificial neural net...
详细信息
ISBN:
(数字)9781665490429
ISBN:
(纸本)9781665490429
Artificial neural network (ANN) has been widely used in automation. However, the vulnerability of ANN under certain attacks poses security threat for critical automation systems. Adding noises to artificial neural network has been shown to be able to improve robustness in previous work. In this work, we propose a new technique to compute the pathwise stochastic gradient estimate with respect to the standard deviation of the Gaussian noise added to each neuron of the ANN. By our proposed technique, the gradient estimation with respect to noise levels is a byproduct of the back propagation algorithm for estimating gradient with respect to synaptic weights in ANN. Thus, the noise level for each neuron can be optimized simultaneously in the processing of training the synaptic weights at nearly no extra computational cost. In numerical experiments, our proposed method can achieve significant performance improvement on robustness of several popular ANN structures under both black box and white box attacks.
暂无评论