Adversarial Constraint Learning for Structured Prediction

by   Hongyu Ren, et al.
Stanford University

Constraint-based learning reduces the burden of collecting labels by having users specify general properties of structured outputs, such as constraints imposed by physical laws. We propose a novel framework for simultaneously learning these constraints and using them for supervision, bypassing the difficulty of using domain expertise to manually specify constraints. Learning requires a black-box simulator of structured outputs, which generates valid labels, but need not model their corresponding inputs or the input-label relationship. At training time, we constrain the model to produce outputs that cannot be distinguished from simulated labels by adversarial training. Providing our framework with a small number of labeled inputs gives rise to a new semi-supervised structured prediction model; we evaluate this model on multiple tasks --- tracking, pose estimation and time series prediction --- and find that it achieves high accuracy with only a small number of labeled inputs. In some cases, no labels are required at all.


Learning Constraints for Structured Prediction Using Rectifier Networks

Various natural language processing tasks are structured prediction prob...

Predict and Constrain: Modeling Cardinality in Deep Structured Prediction

Many machine learning problems require the prediction of multi-dimension...

RankingMatch: Delving into Semi-Supervised Learning with Consistency Regularization and Ranking Loss

Semi-supervised learning (SSL) has played an important role in leveragin...

Query-Adaptive Predictive Inference with Partial Labels

The cost and scarcity of fully supervised labels in statistical machine ...

Adversarial Attack and Defense of Structured Prediction Models

Building an effective adversarial attacker and elaborating on countermea...

Effective and Efficient Data Poisoning in Semi-Supervised Learning

Semi-Supervised Learning (SSL) aims to maximize the benefits of learning...

Simplifying Models with Unlabeled Output Data

We focus on prediction problems with high-dimensional outputs that are s...

Please sign up or login with your details

Forgot password? Click here to reset