Avoiding Your Teacher's Mistakes: Training Neural Networks with Controlled Weak Supervision

Abstract

Training deep neural networks requires massive amounts of training data, but for many tasks only limited labeled data is available. This makes weak supervision attractive, using weak or noisy signals like the output of heuristic methods or user click-through data for training. In a semi-supervised setting, we can use a large set of data with weak labels to pretrain a neural network and then fine-tune the parameters with a small amount of data with true labels. This feels intuitively sub-optimal as these two independent stages leave the model unaware about the varying label quality. What if we could somehow inform the model about the label quality? In this paper, we propose a semi-supervised learning method where we train two neural networks in a multi-task fashion: a "target network" and a "confidence network". The target network is optimized to perform a given task and is trained using a large set of unlabeled data that are weakly annotated. We propose to weight the gradient updates to the target network using the scores provided by the second confidence network, which is trained on a small amount of supervised data. Thus we avoid that the weight updates computed from noisy labels harm the quality of the target network model. We evaluate our learning strategy on two different tasks: document ranking and sentiment classification. The results demonstrate that our approach not only enhances the performance compared to the baselines but also speeds up the learning process from weak labels.

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Distilling the Knowledge in a Neural Network
• FitNets: Hints for Thin Deep Nets
• Training Convolutional Networks with Noisy Labels
• Data Programming: Creating Large Training Sets, Quickly
• HoloClean: Holistic Data Repairs with Probabilistic Inference
• Learning From Noisy Large-Scale Datasets With Minimal Supervision
• TF.Learn: TensorFlow's High-level Module for Distributed Machine Learning
• Learning to Learn from Weak Supervision by Full Supervision
• Socratic Learning: Augmenting Generative Models to Incorporate Latent Subsets in Training Data
• Learning to Attend, Copy, and Generate for Session-Based Query Suggestion

Cited by in corpus (11)

• Image Classification with Deep Learning in the Presence of Noisy Labels: A Survey
• Suppressing Uncertainties for Large-Scale Facial Expression Recognition
• Denoising Multi-Source Weak Supervision for Neural Text Classification
• Learning Not to Learn in the Presence of Noisy Labels
• A Deep Look into Neural Ranking Models for Information Retrieval
• Fidelity-Weighted Learning
• Learning Matching Models with Weak Supervision for Response Selection in Retrieval-based Chatbots
• Towards Theoretical Understanding of Weak Supervision for Information Retrieval
• Recommending Courses in MOOCs for Jobs: An Auto Weak Supervision Approach
• Learning to Rank from Samples of Variable Quality
• Task-Driven Data Verification via Gradient Descent