Abstract:Deep learning algorithms have recently produced state-of-the-art accuracy in many classification tasks, but this success is typically dependent on access to many annotated training examples. For domains without such data, an attractive alternative is to train models with light, or distant supervision. In this paper, we introduce a deep neural network for the Learning from Label Proportion (LLP) setting, in which the training data consist of bags of unlabeled instances with associated label distributions for each bag. We introduce a new regularization layer, Batch Averager, that can be appended to the last layer of any deep neural network to convert it from supervised learning to LLP. This layer can be implemented readily with existing deep learning packages. To further support domains in which the data consist of two conditionally independent feature views (e.g. image and text), we propose a co-training algorithm that iteratively generates pseudo bags and refits the deep LLP model to improve classification accuracy. We demonstrate our models on demographic attribute classification (gender and race/ethnicity), which has many applications in social media analysis, public health, and marketing. We conduct experiments to predict demographics of Twitter users based on their tweets and profile image, without requiring any user-level annotations for training. We find that the deep LLP approach outperforms baselines for both text and image features separately. Additionally, we find that co-training algorithm improves image and text classification by 4% and 8% absolute F1, respectively. Finally, an ensemble of text and image classifiers further improves the absolute F1 measure by 4% on average.

Leveraging Interactive Knowledge And Unlabeled Data In Gender Classification With Co-Training

Gender Classification Based On Evolutionary Extreme Learning Machine

Interactive Gender Inference in Social Media

Inter-training: Exploiting Unlabeled Data in Multi-Classifier Systems

CoTrade: Confident Co-Training With Data Editing.

Advanced people flow monitoring and gender classification: A joint application of YOLO, DeepSORT, and convolutional neural networks

Co-training for Cross-Lingual Sentiment Classification

Speech Emotion Recognition Using an Enhanced Co-Training Algorithm

Exploring Representations from Unlabeled Data with Co-training for Chinese Word Segmentation.

Hessian-regularized Co-Training for Social Activity Recognition.

Co-training for Demographic Classification Using Deep Learning from Label Proportions

Multi-Dimensional Gender Bias Classification

User Gender Classification in Chinese Microblog

Tri-Training: Exploiting Unlabeled Data Using Three Classifiers

Semi-Supervised Regression with Co-Training

Bilingual co-training for sentiment classification of chinese product reviews

Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints

Automatically Inferring Gender Associations from Language

Online User Modeling for Interactive Streaming Image Classification.

Multi-task Learning for Gender and Age Prediction on Chinese Microblog

Text as Image: Learning Transferable Adapter for Multi-Label Classification