Proceedings:
No. 11: AAAI-21 Technical Tracks 11
Volume
Issue:
Proceedings of the AAAI Conference on Artificial Intelligence, 35
Track:
AAAI Technical Track on Machine Learning IV
Downloads:
Abstract:
We present *-CFQ ("star-CFQ"): a suite of large-scale datasets of varying scope based on the CFQ semantic parsing benchmark, designed for principled investigation of the scalability of machine learning systems in a realistic compositional task setting. Using this suite, we conduct a series of experiments investigating the ability of Transformers to benefit from increased training data size under conditions of fixed computational cost. We show that compositional generalization remains a challenge at all training sizes, and we show that increasing the scope of natural language leads to consistently higher error rates, which are only partially offset by increased training data. We further show that while additional training data from a related domain improves the accuracy in data-starved situations, this improvement is limited and diminishes as the distance from the related domain to the target domain increases.
DOI:
10.1609/aaai.v35i11.17195
AAAI
Proceedings of the AAAI Conference on Artificial Intelligence, 35