IEEE Project Abstract

Inspired by the recent successes of deep architecture,the auto-encoder and its variants have been intensively explored on image clustering and classification tasks by learning effective feature representations. Conventional auto-encoder attempts to uncover the data’s intrinsic structure of by constraining the output to be as much identical to the input as possible, which denotes the hidden representation could faithfully reconstruct the input data. One issue arises, however, that such representations might not be optimized for specific tasks, e.g, image classification and clustering, since it compresses not only the discriminative information but also lots of redundant or even noise within data. In other words, not all hidden units would benefit the specific tasks, while partial units are mainly used to represent the task-irrelevant patterns. In this work, a general frame work named Discerning Feature Supported Encoder (DFSE) is proposed which integrates auto-encoder and feature selection together into a unified model. Specifically, the feature selection is adapted to learned hidden-layer features to capture the task-relevant ones from the task-irrelevant ones. Meanwhile, the selected hidden units could in turn encode more discriminability only on selected task-relevant units. To this end, our proposed algorithm can generate more effective image representation by distinguishing the task-relevant features from the -irrelevant ones. Two scenarios of experiments on image classification and clustering are conducted to evaluate our algorithm. Experiments on several benchmarks demonstrate our method can achieve better performance overstate-of-the-art approaches in two scenarios.