Learning Modular Structures That Generalize Out-of-Distribution (Student Abstract)
Arjun Ashok, Chaitanya Devaguptapu, Vineeth Balasubramanian
[AAAI-22] Student Abstract and Poster Program
Abstract:
Out-of-distribution (O.O.D.) generalization remains to be a key challenge for real-world machine learning systems. We describe a method for O.O.D. generalization that, through training, encourages models to only preserve features in the network that are well reused across multiple training domains. Our method combines two complementary neuron-level regularizers with a probabilistic differentiable binary mask over the network, to extract a modular sub-network that achieves better O.O.D. performance than the original network. Preliminary evaluation on two benchmark datasets corroborates the promise of our method.
Sessions where this paper appears
-
Poster Session 2
Fri, February 25 12:45 AM - 2:30 AM (+00:00)
Blue 2
-
Poster Session 9
Sun, February 27 8:45 AM - 10:30 AM (+00:00)
Blue 2