Analyzing machine-learned representations: A natural language case study

Understand

As modern deep networks become more complex, and get closer to human-like capabilities in certain domains, the question arises of how the representations and decision rules they learn compare to the ones in humans.

• In this work, we study representations of sentences in one such artificial system for natural language processing.
• We first present a diagnostic test dataset to examine the degree of abstract composable structure represented.
• Analyzing performance on these diagnostic tests indicates a lack of systematicity in the representations and decision rules, and reveals a set of heuristic strategies.