# Differentiable Compositional Kernel Learning for Gaussian Processes

arXiv: https://arxiv.org/abs/1806.04326 · HTML: https://mlsanity.com/paper/1806.04326/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Jun 2018.

## Built on by (most cited first, 23 of 23)

- [Priors in Bayesian Deep Learning: A Review](https://mlsanity.com/paper/2105.06868/) (2021)
- [Prediction-Oriented Bayesian Active Learning](https://mlsanity.com/paper/2304.08151/) (2023)
- [Contrastive Learning Can Find An Optimal Basis For Approximately View-Invariant Functions](https://mlsanity.com/paper/2210.01883/) (2022)
- [A Primer on Bayesian Neural Networks: Review and Debates](https://mlsanity.com/paper/2309.16314/) (2023)
- [Kernel Identification Through Transformers](https://mlsanity.com/paper/2106.08185/) (2021)
- [Raven's Progressive Matrices Completion with Latent Gaussian Process Priors](https://mlsanity.com/paper/2103.12045/) (2021)
- [Structural Kernel Search via Bayesian Optimization and Symbolical Optimal Transport](https://mlsanity.com/paper/2210.11836/) (2022)
- [Neural auto-designer for enhanced quantum kernels](https://mlsanity.com/paper/2401.11098/) (2024)
- [Batch Bayesian Optimization on Permutations using the Acquisition Weighted Kernel](https://mlsanity.com/paper/2102.13382/) (2021)
- [The Minecraft Kernel: Modelling correlated Gaussian Processes in the Fourier domain](https://mlsanity.com/paper/2103.06950/) (2021)
- [Gaussian Process Latent Variable Modeling for Few-shot Time Series Forecasting](https://mlsanity.com/paper/2212.10306/) (2022)
- [A Self-Supervised Framework for Function Learning and Extrapolation](https://mlsanity.com/paper/2106.07369/) (2021)
- [Modelling Non-Smooth Signals with Complex Spectral Structure](https://mlsanity.com/paper/2203.06997/) (2022)
- [On the Identifiability and Interpretability of Gaussian Process Models](https://mlsanity.com/paper/2310.17023/) (2023)
- [KATO: Knowledge Alignment and Transfer for Transistor Sizing of Different Design and Technology](https://mlsanity.com/paper/2404.14433/) (2024)
- [From Deep Additive Kernel Learning to Last-Layer Bayesian Neural Networks via Induced Prior Approximation](https://mlsanity.com/paper/2502.10540/) (2025)
- [Conditional Deep Gaussian Processes: empirical Bayes hyperdata learning](https://mlsanity.com/paper/2110.00568/) (2021)
- [On Connecting Deep Trigonometric Networks with Deep Gaussian Processes: Covariance, Expressivity, and Neural Tangent Kernel](https://mlsanity.com/paper/2203.07411/) (2022)
- [Volatility Based Kernels and Moving Average Means for Accurate Forecasting with Gaussian Processes](https://mlsanity.com/paper/2207.06544/) (2022)
- [Online structural kernel selection for mobile health](https://mlsanity.com/paper/2107.09949/) (2021)
- [Scalable Deep Basis Kernel Gaussian Processes](https://mlsanity.com/paper/2505.18526/) (2025)
- [Kernel Model Validation: How To Do It, And Why You Should Care](https://mlsanity.com/paper/2509.15244/) (2025)
- [Uncertainty-quantified Pulse Signal Recovery from Facial Video using Regularized Stochastic Interpolants](https://mlsanity.com/paper/2604.10777/) (2026)

## Abstract (arXiv)

The generalization properties of Gaussian processes depend heavily on the choice of kernel, and this choice remains a dark art. We present the Neural Kernel Network (NKN), a flexible family of kernels represented by a neural network. The NKN architecture is based on the composition rules for kernels, so that each unit of the network corresponds to a valid kernel. It can compactly approximate compositional kernel structures such as those used by the Automatic Statistician (Lloyd et al., 2014), but because the architecture is differentiable, it is end-to-end trainable with gradient-based optimization. We show that the NKN is universal for the class of stationary kernels. Empirically we demonstrate pattern discovery and extrapolation abilities of NKN on several tasks that depend crucially on identifying the underlying structure, including time series and texture extrapolation, as well as Bayesian optimization.