Semi-Supervised Learning with Declaratively Specified Entropy Constraints
arXiv:1804.09238
Abstract
We propose a technique for declaratively specifying strategies for semi-supervised learning (SSL). The proposed method can be used to specify ensembles of semi-supervised learning, as well as agreement constraints and entropic regularization constraints between these learners, and can be used to model both well-known heuristics such as co-training and novel domain-specific heuristics. In addition to representing individual SSL heuristics, we show that multiple heuristics can also be automatically combined using Bayesian optimization methods. We show consistent improvements on a suite of well-studied SSL benchmarks, including a new state-of-the-art result on a difficult relation extraction task.
References in corpus (6)
- Semi-Supervised Classification with Graph Convolutional Networks
- Practical Bayesian Optimization of Machine Learning Algorithms
- Snorkel: Rapid Training Data Creation with Weak Supervision
- Revisiting Semi-Supervised Learning with Graph Embeddings
- Distant IE by Bootstrapping Using Lists and Document Structure
- Bootstrapping Distantly Supervised IE using Joint Learning and Small Well-structured Corpora