Does Syntax Need to Grow on Trees? Sources of Hierarchical Inductive Bias in Sequence-to-Sequence Networks
Open Access
- 1 December 2020
- journal article
- research article
- Published by MIT Press in Transactions of the Association for Computational Linguistics
- Vol. 8, 125-140
Abstract
Learners that are exposed to the same training data might generalize differently due to differing inductive biases. In neural network models, inductive biases could in theory arise from any aspect ...Keywords
All Related Versions
This publication has 24 references indexed in Scilit:
- Cognitive science in the era of artificial intelligence: A roadmap for reverse-engineering the infant language-learnerCognition, 2018
- Exploring the Syntactic Abilities of RNNs with Multi-task Learning2017
- Improved Neural Machine Translation with a Syntax-Aware Encoder and Decoder2017
- Poverty of the Stimulus RevisitedCognitive Science, 2011
- Is Structure Dependence an Innate Constraint? New Experimental Evidence From Children's Complex‐Question ProductionCognitive Science, 2008
- A unified architecture for natural language processing2008
- Short-term memory for serial order: A recurrent neural network model.Psychological Review, 2006
- Multitask LearningMachine Learning, 1997
- Finding Structure in TimeCognitive Science, 1990
- Rules and representationsBehavioral and Brain Sciences, 1980