- Diagnostic Classifiers: Revealing how Neural Networks Process Hierarchical Structure
- CEUR Workshop Proceedings
- Article number
- Number of pages
- Document type
- Interfacultary Research Institutes
Faculty of Science (FNWI)
- Institute for Logic, Language and Computation (ILLC)
We investigate how neural networks can be used for hierarchical, compositional semantics. To this end, we define the simple but nontrivial artificial task of processing nested arithmetic expressions and study whether different types of neural networks can learn to add and subtract. We find that recursive neural networks can implement a generalising solution, and we visualise the intermediate steps: projection, summation and squashing. We also show that gated recurrent neural networks, which process the expressions incrementally, perform surprisingly well on this task: they learn to predict the outcome of the arithmetic expressions with reasonable accuracy, although performance deteriorates with increasing length. To analyse what strategy the recurrent network applies, visualisation techniques are less insightful. Therefore, we develop an approach where we formulate and test hypotheses on what strategies these networks might be following. For each hypothesis, we derive predictions about features of the hidden state representations at each time step, and train ’diagnostic classifiers’ to test those predictions. Our results indicate the networks follow a strategy similar to our hypothesised ’incremental strategy’.
- Final publisher version
- Proceedings title: CoCo 2016 : Cognitive Computation : proceedings of the Workshop on Cognitive Computation: Integrating neural
and symbolic approaches 2016, co-located with the 30th Annual Conference on Neural Information Processing Systems (NIPS 2016)
: Barcelona, Spain, December 9, 2016
Place of publication: Aachen
Editors: T.R. Besold, A. Bordes, A. d'Avila Garcez, G. Wayne.
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library, or send a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.