Learning representations of neural network weights given a model zoo is an emerg- ing and challenging area with many potential applications from model inspection, to neural architecture search or knowledge distillation. Recently, an autoencoder trained on a model zoo was able to learn a hyper-representation, which captures intrinsic and extrinsic properties of the models in the zoo. In this work, we ex- tend hyper-representations for generative use to sample new model weights. We propose layer-wise loss normalization which we demonstrate is key to generate high-performing models and several sampling methods based on the topology of hyper-representations. The models generated using our methods are diverse, per- formant and capable to outperform strong baselines as evaluated on several down- stream tasks: initialization, ensemble sampling and transfer learning. Our results indicate the potential of knowledge aggregation from model zoos to new models via hyper-representations thereby paving the avenue for novel research directions.
contribution to scientific community
HSG Profile Area
Advances in Neural Information Processing Systems
Curran Associates, Inc.
Conference on Neural Information Processing Systems