# Search-Guided, Lightly-supervised Training of Structured Prediction Energy Networks

## TL;DR

This paper introduces a new training method for structured prediction energy networks that uses reward-based randomized search instead of labeled data, enabling effective learning with less supervision.

## Contribution

It proposes a novel training approach using truncated randomized search in reward functions for SPENs, reducing reliance on labeled data and improving training efficiency.

## Key findings

- Effective supervision from reward-based search
- Achieves state-of-the-art results in structured prediction
- Reduces need for expensive labeled training data

## Abstract

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.