PACOH: Bayes-Optimal Meta-Learning with PAC-Guarantees

2020-02-13 · ICML Workshop LifelongML 2020 7 · Jonas Rothfuss, Vincent Fortuin, Martin Josifoski, Andreas Krause

Meta-learning can successfully acquire useful inductive biases from data. Yet, its generalization properties to unseen learning tasks are poorly understood. Particularly if the number of meta-training tasks is small, this raises concerns about overfitting. We provide a theoretical analysis using the PAC-Bayesian framework and derive novel generalization bounds for meta-learning. Using these bounds, we develop a class of PAC-optimal meta-learning algorithms with performance guarantees and a principled meta-level regularization. Unlike previous PAC-Bayesian meta-learners, our method results in a standard stochastic optimization problem which can be solved efficiently and scales well. When instantiating our PAC-optimal hyper-posterior (PACOH) with Gaussian processes and Bayesian Neural Networks as base learners, the resulting methods yield state-of-the-art performance, both in terms of predictive accuracy and the quality of uncertainty estimates. Thanks to their principled treatment of uncertainty, our meta-learners can also be successfully employed for sequential decision problems.

📄 PDF Abstract BibTeX arXiv:2002.05551

Code (3)

jonasrothfuss/meta_learning_pacoh 공식 구현 pytorch

jonasrothfuss/pacoh_nn tf

yilun-wu/MetaL-Benchmark

Tasks

Gaussian ProcessesGeneralization BoundsMeta-Learning Stochastic Optimization

Scalable PAC-Bayesian Meta-Learning via the PAC-Optimal Hyper-Posterior: From Theory to Practice

2022-11-14 · Jonas Rothfuss, Martin Josifoski, Vincent Fortuin, Andreas Krause

Meta-Learning aims to speed up the learning process on new tasks by acquiring useful inductive biases from datasets of related learning tasks. While, in practice, the number of related tasks available is often small, mos…

Gaussian Processes Meta-Learning Stochastic Optimization

Bayesian-Boosted MetaLoc: Efficient Training and Guaranteed Generalization for Indoor Localization

2023-08-28 · Dongze Wu, Jun Gao, Feng Yin

Existing localization approaches utilizing environment-specific channel state information (CSI) excel under specific environment but struggle to generalize across varied environments. This challenge becomes even more pro…

Indoor Localization Meta-Learning

Meta-Learning Reliable Priors in the Function Space

2021-06-06 · NeurIPS 2021 12 · Jonas Rothfuss, Dominique Heyn, Jinfan Chen, Andreas Krause

When data are scarce meta-learning can improve a learner's accuracy by harnessing previous experience from related learning tasks. However, existing methods have unreliable uncertainty estimates which are often overconfi…

Bayesian Optimization Decision Making Meta-Learning Sequential Decision Making+1

Data-Efficient Task Generalization via Probabilistic Model-based Meta Reinforcement Learning

2023-11-13 · Arjun Bhardwaj, Jonas Rothfuss, Bhavya Sukhija, Yarden As 외

We introduce PACOH-RL, a novel model-based Meta-Reinforcement Learning (Meta-RL) algorithm designed to efficiently adapt control policies to changing dynamics. PACOH-RL meta-learns priors for the dynamics model, allowing…

Meta-Learning Meta Reinforcement Learning reinforcement-learning Uncertainty Quantification

ALPaCA vs. GP-based Prior Learning: A Comparison between two Bayesian Meta-Learning Algorithms

2020-10-15 · YiLun Wu

Meta-learning or few-shot learning, has been successfully applied in a wide range of domains from computer vision to reinforcement learning. Among the many frameworks proposed for meta-learning, bayesian methods are part…

Few-Shot Learning Meta-Learning Reinforcement Learning (RL)