Learning compositionally through attentive guidance

Abstract

While neural network models have been successfully applied to domains that require substantial generalisation skills, recent studies have implied that they struggle when solving the task they are trained on requires inferring its underlying compositional structure. In this paper, we introduce Attentive Guidance, a mechanism to direct a sequence to sequence model equipped with attention to find more compositional solutions. We test it on two tasks, devised precisely to assess the compositional capabilities of neural models, and we show that vanilla sequence to sequence models with attention overfit the training distribution, while the guided versions come up with compositional solutions that fit the training and testing distributions almost equally well. Moreover, the learned solutions generalise even in cases where the training and testing distributions strongly diverge. In this way, we demonstrate that sequence to sequence models are capable of finding compositional solutions without requiring extra components. These results helps to disentangle the causes for the lack of systematic compositionality in neural networks, which can in turn fuel future work.

References in corpus (8)

• Sequence to Sequence Learning with Neural Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Categorical Reparameterization with Gumbel-Softmax
• Neural Programmer-Interpreters
• Modular Multitask Reinforcement Learning with Policy Sketches
• Making Neural Programming Architectures Generalize via Recursion
• Memorize or generalize? Searching for a compositional RNN in a haystack
• Exploring Human-like Attention Supervision in Visual Question Answering

Cited by in corpus (6)

• Compositionality decomposed: how do neural networks generalise?
• The Neural Data Router: Adaptive Control Flow in Transformers Improves Systematic Generalization
• Assessing Phrasal Representation and Composition in Transformers
• On the Interplay Between Fine-tuning and Composition in Transformers
• Can Transformers Jump Around Right in Natural Language? Assessing Performance Transfer from SCAN
• On learning an interpreted language with recurrent models