Learning to rank with hidden measures – We propose an ensemble model to estimate the importance of a ranking by combining the data of two different classes. We formulate the method as an optimization problem over the learning function, and then build an ensemble of multiple models. Our performance evaluation shows that the ensemble model achieves reasonable performance compared with the state-of-the-art ensemble learning methods for rank and ranking.
To better understand how such a system works, we firstly explore the problem of learning to perform MorphMan with an adaptive learning approach based on the recognition of morphological patterns from the input data. We develop the first model trained to learn MorphMan, by proposing an algorithm of learning an adaptive learning algorithm that learns a morphological model using its data. We show that the adaptive learning algorithm is able to recognize morphological patterns that are similar to the output of the MorphMan algorithm, in the sense that the learned model has a common representation and a common morphological form. Experiments on real-world morphological data have shown that our approach is superior to the state of the art.
An Interactive Spatial Data Segmentation System
Robust Low-rank Spatial Pyramid Modeling with Missing Labels using Generative Adversarial Network
Learning to rank with hidden measures
A Deep Neural Network based on Energy Minimization
MorphMan: A System for Morph RecognitionTo better understand how such a system works, we firstly explore the problem of learning to perform MorphMan with an adaptive learning approach based on the recognition of morphological patterns from the input data. We develop the first model trained to learn MorphMan, by proposing an algorithm of learning an adaptive learning algorithm that learns a morphological model using its data. We show that the adaptive learning algorithm is able to recognize morphological patterns that are similar to the output of the MorphMan algorithm, in the sense that the learned model has a common representation and a common morphological form. Experiments on real-world morphological data have shown that our approach is superior to the state of the art.