Timezone: »

Search-Guided, Lightly-Supervised Training of Structured Prediction Energy Networks
Amirmohammad Rooshenas · Dongxu Zhang · Gopal Sharma · Andrew McCallum

Thu Dec 12 05:00 PM -- 07:00 PM (PST) @ East Exhibition Hall B + C #63

In structured output prediction tasks, labeling ground-truth training output is often expensive. However, for many tasks, even when the true output is unknown, we can evaluate predictions using a scalar reward function, which may be easily assembled from human knowledge or non-differentiable pipelines. But searching through the entire output space to find the best output with respect to this reward function is typically intractable. In this paper, we instead use efficient truncated randomized search in this reward function to train structured prediction energy networks (SPENs), which provide efficient test-time inference using gradient-based search on a smooth, learned representation of the score landscape, and have previously yielded state-of-the-art results in structured prediction. In particular, this truncated randomized search in the reward function yields previously unknown local improvements, providing effective supervision to SPENs, avoiding their traditional need for labeled training data.

Author Information

Amirmohammad Rooshenas (University of Massachusetts Amherst)
Dongxu Zhang (University of Massachusetts Amherst)
Gopal Sharma (University of Massachusetts Amherst)
Andrew McCallum (UMass Amherst)

More from the Same Authors