CLAILGApr 14, 2017

Optimizing Differentiable Relaxations of Coreference Evaluation Metrics

arXiv:1704.04451v322 citations
Originality Highly original
AI Analysis

This addresses a bottleneck in coreference resolution for NLP researchers, offering a more efficient alternative to reinforcement learning.

The paper tackles the problem of optimizing coreference evaluation metrics, which are non-differentiable, by proposing a differentiable relaxation that enables gradient-based optimization, resulting in a substantial performance gain over competitive neural systems.

Coreference evaluation metrics are hard to optimize directly as they are non-differentiable functions, not easily decomposable into elementary decisions. Consequently, most approaches optimize objectives only indirectly related to the end goal, resulting in suboptimal performance. Instead, we propose a differentiable relaxation that lends itself to gradient-based optimisation, thus bypassing the need for reinforcement learning or heuristic modification of cross-entropy. We show that by modifying the training objective of a competitive neural coreference system, we obtain a substantial gain in performance. This suggests that our approach can be regarded as a viable alternative to using reinforcement learning or more computationally expensive imitation learning.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes