Timezone: »

 
Poster
Invariant Rationalization
Shiyu Chang · Yang Zhang · Mo Yu · Tommi Jaakkola

Thu Jul 16 06:00 AM -- 06:45 AM & Thu Jul 16 05:00 PM -- 05:45 PM (PDT) @
in Poster Session 41 »

Selective rationalization improves neural network interpretability by identifying a small subset of input features — the rationale — that best explains or supports the prediction. A typical rationalization criterion, i.e. maximum mutual information (MMI), finds the rationale that maximizes the prediction performance based only on the rationale. However, MMI can be problematic because it picks up spurious correlations between the input features and the output. Instead, we introduce a game-theoretic invariant rationalization criterion where the rationales are constrained to enable the same predictor to be optimal across different environments. We show both theoretically and empirically that the proposed rationales can rule out spurious correlations and generalize better to different test scenarios. The resulting explanations also align better with human judgments. Our implementations are publicly available at https://github.com/code-terminator/invariant_rationalization.

Keywords: Accountability, Transparency and Interpretability · Natural Language Processing / Dialogue

Author Information

Shiyu Chang (MIT-IBM Watson AI Lab)
Yang Zhang (MIT-IBM Watson AI Lab)
Mo Yu (IBM T. J. Watson)
Tommi Jaakkola (MIT)

More from the Same Authors