Semi-Parametric Inducing Point Networks and Neural Processes

• Published in: arXiv.org 2023-03
• Main Authors: Rastogi, Richa, Schiff, Yair, Hacohen, Alon, Li, Zhaozhi, Lee, Ian, Deng, Yuntian, Sabuncu, Mert R, Kuleshov, Volodymyr
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Computational efficiency; Computer architecture; Computing costs; Inference; Machine learning; Neural networks; Training
• ISSN: 2331-8422

We introduce semi-parametric inducing point networks (SPIN), a general-purpose architecture that can query the training set at inference time in a compute-efficient manner. Semi-parametric architectures are typically more compact than parametric models, but their computational complexity is often quadratic. In contrast, SPIN attains linear complexity via a cross-attention mechanism between datapoints inspired by inducing point methods. Querying large training sets can be particularly useful in meta-learning, as it unlocks additional training signal, but often exceeds the scaling limits of existing models. We use SPIN as the basis of the Inducing Point Neural Process, a probabilistic model which supports large contexts in meta-learning and achieves high accuracy where existing models fail. In our experiments, SPIN reduces memory requirements, improves accuracy across a range of meta-learning tasks, and improves state-of-the-art performance on an important practical problem, genotype imputation.