Timezone: »

[S12] Efficient Decompositional Rule Extraction for Deep Neural Networks
Mateo Espinosa Zarlenga · Mateja Jamnik

Tue Dec 14 01:50 PM -- 01:53 PM (PST) @
Event URL: https://openreview.net/forum?id=Mla9xdACIvB »

In recent years, there has been significant work on increasing both interpretability and debuggability of a Deep Neural Network (DNN) by extracting a rule-based model that approximates its decision boundary. Nevertheless, current DNN rule extraction methods that consider a DNN's latent space when extracting rules, known as decompositional algorithms, are either restricted to single-layer DNNs or intractable as the size of the DNN or data grows. In this paper, we address these limitations by introducing ECLAIRE, a novel polynomial-time rule extraction algorithm capable of scaling to both large DNN architectures and large training datasets. We evaluate ECLAIRE on a wide variety of tasks, ranging from breast cancer prognosis to particle detection, and show that it consistently extracts more accurate and comprehensible rule sets than the current state-of-the-art methods while using orders of magnitude less computational resources. We make all of our methods available, including a rule set visualisation interface, through the open-source REMIX library.

Author Information

Mateo Espinosa Zarlenga (University of Cambridge)
Mateja Jamnik (University of Cambridge)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors