Timezone: »

Neural Interaction Transparency (NIT): Disentangling Learned Interactions for Improved Interpretability
Michael Tsang · Hanpeng Liu · Sanjay Purushotham · Pavankumar Murali · Yan Liu

Tue Dec 04 07:45 AM -- 09:45 AM (PST) @ Room 210 #83
Neural networks are known to model statistical interactions, but they entangle the interactions at intermediate hidden layers for shared representation learning. We propose a framework, Neural Interaction Transparency (NIT), that disentangles the shared learning across different interactions to obtain their intrinsic lower-order and interpretable structure. This is done through a novel regularizer that directly penalizes interaction order. We show that disentangling interactions reduces a feedforward neural network to a generalized additive model with interactions, which can lead to transparent models that perform comparably to the state-of-the-art models. NIT is also flexible and efficient; it can learn generalized additive models with maximum $K$-order interactions by training only $O(1)$ models.

Author Information

Michael Tsang (University of Southern California)
Hanpeng Liu (University of Southern California)
Sanjay Purushotham (University of Maryland Baltimore County)
Pavankumar Murali (IBM)
Yan Liu (DiDi/University of Southern California)

More from the Same Authors