Tractable Semi-supervised Learning of Complex Structured Prediction Models

  title={Tractable Semi-supervised Learning of Complex Structured Prediction Models},
  author={Kai-Wei Chang and S. Sundararajan and S. Keerthi},
Semi-supervised learning has been widely studied in the literature. However, most previous works assume that the output structure is simple enough to allow the direct use of tractable inference/learning algorithms (e.g., binary label or linear chain). Therefore, these methods cannot be applied to problems with complex structure. In this paper, we propose an approximate semi-supervised learning method that uses piecewise training for estimating the model weights and a dual decomposition approach… Expand
10 Citations
A Simple Label Switching Algorithm for Semisupervised Structural SVMs
  • 3
Large Margin Semi-supervised Structured Output Learning
  • 1
  • PDF
Structural Learning with Amortized Inference
  • 11
  • PDF
A Structured Prediction Approach for Missing Value Imputation
  • PDF
An Integer Linear Programming Framework for Mining Constraints from Data
  • PDF
Efficient Learning and Inference for Big and Complex Data
  • Kai-Wei Chang
  • PDF
Men Also Like Shopping: Reducing Gender Bias Amplification using Corpus-level Constraints
  • 329
  • PDF
Target Language-Aware Constrained Inference for Cross-lingual Dependency Parsing
  • 10
  • PDF
Efficient Algorithms for Global Inference in Internet Marketplaces
  • PDF


Structured Learning with Approximate Inference
  • 148
  • PDF
Piecewise training for structured prediction
  • 57
  • Highly Influential
  • PDF
Deterministic Annealing for Semi-Supervised Structured Output Learning
  • 12
  • PDF
Structured learning with constrained conditional models
  • 111
  • Highly Influential
  • PDF
Learning Efficiently with Approximate Inference via Dual Losses
  • 85
  • PDF
Efficient Learning of Structured Predictors in General Graphical Models
  • 6
Efficient Decomposed Learning for Structured Prediction
  • 31
  • PDF
Semi-supervised Multi-label Classification - A Simultaneous Large-Margin, Subspace Learning Approach
  • 27
  • PDF
Semi-supervised learning for structured output variables
  • 75
  • PDF
Generalized Expectation Criteria for Semi-Supervised Learning with Weakly Labeled Data
  • 180
  • PDF