Creating labeled training sets has become one of the major roadblocks in machine learning. To address this, recent Weak Supervision (WS) frameworks synthesize training labels from multiple potentially noisy supervision sources. However, existing frameworks are restricted to supervision sources that share the same output space as the target task. To extend the scope of usable sources, we formulate Weak Indirect Supervision (WIS), a new research problem for automatically synthesizing training labels based on indirect supervision sources that have different output label spaces. To overcome the challenge of mismatched output spaces, we develop a probabilistic modeling approach, PLRM, which uses user-provided label relations to model and leverage indirect supervision sources. Moreover, we provide a theoretically-principled test of the distinguishability of PLRM for unseen labels, along with an generalization bound. On both image and text classification tasks as well as an industrial advertising application, we demonstrate the advantages of PLRM by outperforming baselines by a margin of 2%-9%.

It is a usual practice to ignore any structural information underlying classes in multi-class classification. In this paper, we propose a graph convolutional network (GCN) augmented neural network classifier to exploit a known, underlying graph structure of labels. The proposed approach resembles an (approximate) inference procedure in, for instance, a conditional random field (CRF). We evaluate the proposed approach on document classification and object recognition and report both accuracies and graph-theoretic metrics that correspond to the consistency of the model's prediction. The experiment results reveal that the proposed model outperforms a baseline method which ignores the graph structures of a label space in terms of graph-theoretic metrics.

Feature Selection plays an important role in machine learning and data mining, and it is often applied as a data pre-processing step. This task can speed up learning algorithms and sometimes improve their performance. In multi-label learning, label dependence is considered another aspect that can contribute to improve learning performance. A replicable and wide systematic review performed by us corroborates this idea. Based on this information, it is believed that considering label dependence during feature selection can lead to better learning performance. The hypothesis of this work is that multi-label feature selection algorithms that consider label dependence will perform better than the ones that disregard it. To this end, we propose multi-label feature selection algorithms that take into account label relations. These algorithms were experimentally compared to the standard approach for feature selection, showing good performance in terms of feature reduction and predictability of the classifiers built using the selected features.