Practical equivariances via relational conditional neural processes

[PDF] arxiv.org

Amortized probabilistic conditioning for optimization, simulation and inference

PE Chang, N Loka, D Huang, U Remes, S Kaski… - arXiv preprint arXiv …, 2024 - arxiv.org
Amortized meta-learning methods based on pre-training have propelled fields like natural
language processing and vision. Transformer-based neural processes and their variants are …
被引用次数：1 相关文章 所有 2 个版本
[PDF] arxiv.org

Approximately Equivariant Neural Processes

M Ashman, C Diaconu, A Weller, W Bruinsma… - arXiv preprint arXiv …, 2024 - arxiv.org
Equivariant deep learning architectures exploit symmetries in learning problems to improve
the sample efficiency of neural-network-based models and their ability to generalise …
相关文章
[PDF] arxiv.org

Noise-Aware Differentially Private Regression via Meta-Learning

O Räisä, S Markou, M Ashman, WP Bruinsma… - arXiv preprint arXiv …, 2024 - arxiv.org
Many high-stakes applications require machine learning models that protect user privacy
and provide well-calibrated, accurate predictions. While Differential Privacy (DP) is the gold …
相关文章 所有 2 个版本
[PDF] arxiv.org

Amortized Bayesian Experimental Design for Decision-Making

D Huang, Y Guo, L Acerbi, S Kaski - arXiv preprint arXiv:2411.02064, 2024 - arxiv.org
Many critical decisions, such as personalized medical diagnoses and product pricing, are
made based on insights gained from designing, observing, and analyzing a series of …
相关文章 所有 4 个版本