`

Timezone: »

 
Poster
Firefly Neural Architecture Descent: a General Approach for Growing Neural Networks
Lemeng Wu · Bo Liu · Peter Stone · Qiang Liu

Tue Dec 08 09:00 AM -- 11:00 AM (PST) @ Poster Session 1 #295
in Poster Session 2 »

We propose firefly neural architecture descent, a general framework for progressively and dynamically growing neural networks to jointly optimize the networks' parameters and architectures. Our method works in a steepest descent fashion, which iteratively finds the best network within a functional neighborhood of the original network that includes a diverse set of candidate network structures. By using Taylor approximation, the optimal network structure in the neighborhood can be found with a greedy selection procedure. We show that firefly descent can flexibly grow networks both wider and deeper, and can be applied to learn accurate but resource-efficient neural architectures that avoid catastrophic forgetting in continual learning. Empirically, firefly descent achieves promising results on both neural architecture search and continual learning. In particular, on a challenging continual image classification task, it learns networks that are smaller in size but have higher average accuracy than those learned by the state-of-the-art methods.

[ Paper ]

Author Information

Lemeng Wu (UT Austin)
Bo Liu (University of Texas at Austin)
Peter Stone (The University of Texas at Austin, Sony AI)
Qiang Liu (UT Austin)

More from the Same Authors