# Meta-Learning with Shared Amortized Variational Inference

arXiv: https://arxiv.org/abs/2008.12037 · HTML: https://mlsanity.com/paper/2008.12037/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Aug 2020.

## Built on by (most cited first, 14 of 14)

- [Distribution Transformers: Fast Approximate Bayesian Inference With On-The-Fly Prior Adaptation](https://mlsanity.com/paper/2502.02463/) (2025)
- [Effortless, Simulation-Efficient Bayesian Inference using Tabular Foundation Models](https://mlsanity.com/paper/2504.17660/) (2025)
- [Transductive Decoupled Variational Inference for Few-Shot Classification](https://mlsanity.com/paper/2208.10559/) (2022)
- [Foundation Posteriors for Approximate Probabilistic Inference](https://mlsanity.com/paper/2205.09735/) (2022)
- [BECLR: Batch Enhanced Contrastive Few-Shot Learning](https://mlsanity.com/paper/2402.02444/) (2024)
- [ADAVI: Automatic Dual Amortized Variational Inference Applied To Pyramidal Bayesian Models](https://mlsanity.com/paper/2106.12248/) (2021)
- [Horizon-Free and Variance-Dependent Reinforcement Learning for Latent Markov Decision Processes](https://mlsanity.com/paper/2210.11604/) (2022)
- [Towards Enabling Meta-Learning from Target Models](https://mlsanity.com/paper/2104.03736/) (2021)
- [PAVI: Plate-Amortized Variational Inference](https://mlsanity.com/paper/2206.05111/) (2022)
- [Neural Variational Dropout Processes](https://mlsanity.com/paper/2510.19425/) (2025)
- [Meta-Learning an Inference Algorithm for Probabilistic Programs](https://mlsanity.com/paper/2103.00737/) (2021)
- [PAVI: Plate-Amortized Variational Inference](https://mlsanity.com/paper/2308.16022/) (2023)
- [Federated Learning via Meta-Variational Dropout](https://mlsanity.com/paper/2510.20225/) (2025)
- [Variational Rectification Inference for Learning with Noisy Labels](https://mlsanity.com/paper/2603.17255/) (2026)

## Abstract (arXiv)

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.