DeepAI AI Chat
Log In Sign Up

Estimation from Indirect Supervision with Linear Moments

by   Aditi Raghunathan, et al.

In structured prediction problems where we have indirect supervision of the output, maximum marginal likelihood faces two computational obstacles: non-convexity of the objective and intractability of even a single gradient computation. In this paper, we bypass both obstacles for a class of what we call linear indirectly-supervised problems. Our approach is simple: we solve a linear system to estimate sufficient statistics of the model, which we then use to estimate parameters via convex optimization. We analyze the statistical properties of our approach and show empirically that it is effective in two settings: learning with local privacy constraints and learning from low-cost count-based annotations.


page 1

page 2

page 3

page 4


Learning from Indirect Observations

Weakly-supervised learning is a paradigm for alleviating the scarcity of...

Probable convexity and its application to Correlated Topic Models

Non-convex optimization problems often arise from probabilistic modeling...

Learnability with Indirect Supervision Signals

Learning from indirect supervision signals is important in real-world AI...

Improving Distantly Supervised Relation Extraction by Natural Language Inference

To reduce human annotations for relation extraction (RE) tasks, distantl...

Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction?

Two key obstacles in biomedical relation extraction (RE) are the scarcit...

Summarization as Indirect Supervision for Relation Extraction

Relation extraction (RE) models have been challenged by their reliance o...