Bayesian Model-Agnostic Meta-Learning

By Taesup Kim, Jaesik Yoon, Ousmane Amadou Dia, Sungwoong Kim, Yoshua Bengio, Sungjin Ahn. 2018

548 citations recorded by the source corpus.

Learning to infer Bayesian posterior from a few-shot dataset is an important step towards robust meta-learning due to the model uncertainty inherent in the problem. In this paper, we propose a novel Bayesian model-agnostic meta-learning method. The proposed method combines scalable gradient-based meta-learning with nonparametric variational inference in a principled probabilistic framework. During fast adaptation, the method is capable of learning complex uncertainty structure beyond a point estimate or a simple Gaussian approximation. In addition, a robust Bayesian meta-update mechanism with a new meta-loss prevents overfitting during meta-update. Remaining an efficient gradient-based meta-learner, the method is also model-agnostic and simple to implement. Experiment results show the accuracy and robustness of the proposed method in various tasks: sinusoidal regression, image classification, active learning, and reinforcement learning.

Subfields: Artificial Intelligence, Probabilistic & Bayesian Methods, Meta & Continual Learning

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

In Probabilistic & Bayesian Methods

In Meta & Continual Learning

Citations in the corpus

Meta-Learning Probabilistic Inference for Prediction (2018, background): “(cid:12)(cid:12) This is an example of semi-amortized inference (Kim et al., 2018b), as the only shared inference parameters are the initialization and learning rate, and optimization is required for each task (albeit only for one step).”
Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples (2019, background): “Some bear similarity with MAML and correspond to a meta-learner trained to quickly adapt its parameters to various novel tasks (Ravi & Larochelle, 2017; Munkhdalai & Yu, 2017; Rusu et al., 2019; Yoon et al., 2018).”
Meta-Learning with Latent Embedding Optimization (2018, methodology): “[16] and [7] propose probabilistic extensions to MAML that are trained using a variational approximation using simple posteriors.”
A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms (2019, neutral): “, 2017), and its related algorithms (Grant et al., 2018; Kim et al., 2018; Finn et al., 2018).”
Towards Continual Reinforcement Learning: A Review and Perspectives (2020, background): “There have been a number of recent improvements on this topic of meta-optimization in multi-task RL (Nichol and Schulman, 2018; Kim et al., 2018; Rothfuss et al., 2018; Flennerhag et al., 2018; Nagabandi et al., 2018; Mendonca et al., 2019; Finn et al., 2019; Lin et al., 2020; Berseth et al., 2021; ”
Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variables (2019, methodology): “Hierarchical Bayesian models have been used to model few-shot learning (Fei-Fei et al., 2003; Tenenbaum, 1999), including approaches that perform gradient-based adaptation (Grant et al., 2018; Yoon et al., 2018).”
Meta-Learning in Neural Networks: A Survey (2020, methodology): “Bayesian MAML [251] improves on this model by using a Bayesian ensemble approach that allows non-Gaussian posteriors over θ, and later work removes the need for costly ensembles [45], [252].”
Meta-Curvature (2019, background): “for continuous adaptation [1], better credit assignment [37], and robustness [15].”
Generalizing from a Few Examples (2019, methodology): “Examples include one-shot imitation [147], multi-armed bandits [33], visual navigation [37], and continuous control [156].”
An Ensemble of Epoch-Wise Empirical Bayes for Few-Shot Learning (2019, background): “proposed Bayesian MAML (BMAML) that trains multiple instances of base-model to reduce mete-level overfitting [80].”
The effects of negative adaptation in Model-Agnostic Meta-Learning (2018, unknown): “This algorithm has benefited from a lot of extensions, including ways to incorporate uncertainty estimates [8, 5].”
Reward Shaping via Meta-Learning (2019, methodology): “In light of this and recent works (Grant et al., 2018; Yoon et al., 2018; Finn et al., 2018), we call Φ( s ) the potential function prior .”
Transformers Can Do Bayesian Inference (2021, unknown): “013 BMAML (Yoon et al., 2018) 0 .”
Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Visuomotor Policies (2018, background): “Reinforcement Learning, [76, 50, 71, 45] and its variants like Meta-RL [20, 27, 54, 21, 40, 74, 17, 49] or its sister fields such as imitation learning [1, 26, 22, 89, 37, 63], commonly focus on the last part of the end-to-end active task pipeline: how to choose an action given a “state” from the wo”
Mid-Level Visual Representations Improve Generalization and Sample Efficiency for Learning Active Tasks (2018, background): “Reinforcement Learning , [46] and its variants like Meta-RL [10, 14, 30, 12, 21] or its sister ﬁelds such as Imitation Learning [2], commonly focus on the last part of the end-to-end active task pipeline: how to choose an action given a state from the world.”
Sharp-MAML: Sharpness-Aware Model-Agnostic Meta Learning (2022, methodology): “As baselines, we use MAML (Finn et al., 2017), Matching Nets (Vinyals et al., 2016), CAVIA (Zintgraf et al., 2019b), Reptile (Nichol & Schulman, 2018), FOMAML (Nichol & Schulman, 2018), and BMAML (Yoon et al., 2018).”
Stein Variational Gradient Descent as Moment Matching (2018, background): “ras, 2018), reinforcement learning (Liu et al., 2017; Haarnoja et al., 2017), learning deep probabilistic models (Wang & Liu, 2016; Pu et al., 2017) and Bayesian meta learning (Feng et al., 2017; Kim et al., 2018). However, the theoretical properties of SVGD are still largely unexplored. The onl”