Meta-Learning with Shared Amortized Variational Inference
By Ekaterina Iakovleva, Jakob Verbeek, Alahari Karteek. 2020
26 citations recorded by the source corpus.
We propose a novel amortized variational inference scheme for an empirical Bayes meta-learning model, where model parameters are treated as latent variables. We learn the prior distribution over model parameters conditioned on limited training data using a variational autoencoder approach. Our framework proposes sharing the same amortized inference network between the conditional prior and variational posterior distributions over the model parameters. While the posterior leverages both the labeled support and query data, the conditional prior is based only on the labeled support data. We show that in earlier work, relying on Monte-Carlo approximation, the conditional prior collapses to a Dirac delta function. In contrast, our variational approach prevents this collapse and preserves uncertainty over the model parameters. We evaluate our approach on the miniImageNet, CIFAR-FS and FC100 datasets, and present results demonstrating its advantages over previous work.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
Citations in the corpus
- Foundation Posteriors for Approximate Probabilistic Inference (2022, neutral): “Meta-Inference Meta-learning has been used to do inference amortized over both queries and a family of generative models [16, 30, 36].”
- Meta-Learning an Inference Algorithm for Probabilistic Programs (2021, background): “…also features in the work on stochastic variational inference where a variational distribution receives information about each inference task in the form of its dataset of observations and is trained with a collection of datasets (Wu et al., 2020; Gordon et al., 2019; Iakovleva et al., 2020).”