Neural Parameter Allocation Search

Neural parameter allocation search (NPAS) automates parameter sharing by obtaining weights for a network given an arbitrary, fixed parameter budget. Prior work has two major drawbacks we aim to address. First, there is a disconnect in the sharing pattern between the search and training steps, where weights are warped for layers of different sizes during the search to measure similarity, but not during training, resulting in reduced performance. To address this, we generate layer weights by learning to compose sets of SuperWeights, which represent a group of trainable parameters. These SuperWeights are created to be large enough so they can be used to represent any layer in the network, but small enough that they are computationally efficient. The second drawback we address is the method of measuring similarity between shared parameters. Whereas prior work compared the weights themselves, we argue this does not take into account the amount of conflict between the shared weights. Instead, we use gradient information to identify layers with shared weights that wish to diverge from each other. We demonstrate that our SuperWeight Networks consistently boost performance over the state-of-the-art on the ImageNet and CIFAR datasets in the NPAS setting. We further show that our approach can generate parameters for many network architectures using the same set of weights. This enables us to support tasks like efficient ensembling and anytime prediction, outperforming fully-parameterized ensembles with 17% fewer parameters 1 . * Equal Contribution, work done while Soren was at Boston University 1 Code available: https://github.com/piotr-teterwak/ SuperWeights a.) Prior Work b.) Ours (SuperWeights) Bilinear Warping Layer Representation Layer Weights No Warping Gradient-based SuperWeight Representation Layer Weights from SuperWeights SSN[23] Templates warped for small and large layers Non-warped SuperWeight templates (Section 2.1) for small and large layers Non-warped SuperWeight templates (Section 2.1) for large layers