Open intent classification is a challenging task in dialogue systems. On the one hand, it should ensure the quality of known intent identification. On the other hand, it needs to detect the open (unknown) intent without prior knowledge. Current models are limited in finding the appropriate decision boundary to balance the performances of both known intents and the open intent. In this paper, we propose a post-processing method to learn the adaptive decision boundary (ADB) for open intent classification. We first utilize the labeled known intent samples to pre-train the model. Then, we automatically learn the adaptive spherical decision boundary for each known class with the aid of well-trained features. Specifically, we propose a new loss function to balance both the empirical risk and the open space risk. Our method does not need open intent samples and is free from modifying the model architecture. Moreover, our approach is surprisingly insensitive with less labeled data and fewer known intents. Extensive experiments on three benchmark datasets show that our method yields significant improvements compared with the state-of-the-art methods.
Using an appropriate control method, linear ultrasonic motors can be used in applications requiring high position accuracy. In this paper, a closed loop PI control system is designed to achieve high position accuracy during the control of a two-DOF stage driven by linear ultrasonic motors. Two ultrasonic motors are mounted on the stage to generate motion in two orthogonal directions. The PI control algorithm is used to increase the stability and accuracy of position control. The x-axis mover covers 30 mm forward and backward in less than 0.3 s settling time and the y-axis mover in less than 0.4 s. Experimental results denote that the control strategy proposed in this paper appears to have high efficiency, quick response, and high accuracy.
Eccentricity is one of the frequent faults of induction motors, and it may cause rub between the rotor and the stator. Early detection of significant rub from pure eccentricity can prolong the lifespan of induction motors. This paper is devoted to such mixed-fault diagnosis: eccentricity plus rub fault. The continuous wavelet transform (CWT) is employed to analyze vibration signals obtained from the motor body. An improved continuous wavelet transform was proposed to alleviate the frequency aliasing. Experimental results show that the proposed method can effectively distinguish two types of faults, single-fault of eccentricity and mixed-fault of eccentricity plus rub.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.