Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
Efficient Bayesian Inference of Sigmoidal Gaussian Cox Processes
Christian Donner, Manfred Opper
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use sigmoid link instead of squared link (mean-field vs Lloyd et al.) | Gaussian Cox process variational inference · 1D synthetic, N=453 | log expected test likelihood: 710.43 -> 720.81 | Improved |
| use sigmoid link instead of squared link | Gaussian Cox process variational inference · 1D synthetic, N=47 | log expected test likelihood: -44.26 -> -43.43 | No clear effect |
| use sigmoid link instead of squared link | Gaussian Cox process variational inference · 1D synthetic, N=4652 | log expected test likelihood: 17496.75 -> 17497.31 | No clear effect |
| use sigmoid link instead of squared link | Gaussian Cox process variational inference · 1D synthetic, N=453 | RMSE: 2.11 -> 0.97 | Improved |
| use sigmoid mean-field instead of squared link | Gaussian Cox process variational inference · 1D synthetic, N=4652 | runtime (s): 0.79 -> 0.41 | Improved |
| use sigmoid mean-field instead of squared link | Gaussian Cox process variational inference · 1D synthetic, N=453 | runtime (s): 0.23 -> 0.50 | Hurt |
| use mean-field or Laplace instead of variational Gauss for sigmoidal Cox process | Sigmoidal Gaussian Cox process · 1D generative-model example | runtime (s): 80 -> 0.09 | Improved |
| use approximate inference instead of MCMC sampler | Sigmoidal Gaussian Cox process · 1D generative-model data (10 datasets) | log expected test likelihood: 875.5 -> 686.2 | Hurt |
| increase number of inducing points beyond about 49 (ablation) | Mean field, Laplace, variational Gauss · 1D synthetic | log expected test likelihood | No clear effect |
| increase number of integration points (ablation) | Mean field, Laplace, variational Gauss · 2D synthetic (10x10 inducing) | log expected test likelihood | No clear effect |
| warm-start by first fitting with 1000 integration points then refitting with target number (ablation) | Sigmoidal mean-field · Mouse neuron and Porto taxi datasets | runtime / test likelihood | Improved |
| use sigmoid link instead of squared link | Gaussian Cox process variational inference · Mouse neuron 2D data | log expected test likelihood | Hurt |
Closest papers by meaning
- Fast Bayesian Intensity Estimation for the Permanental Process (ICML 2017 · similarity 0.93)
- Structured Variational Inference in Continuous Cox Process Models (NeurIPS 2019 · similarity 0.92)
- Fast Bayesian Inference for Gaussian Cox Processes via Path Integral Formulation (NeurIPS 2021 · similarity 0.91)
- Bayesian Optimization through Gaussian Cox Process Models for Spatio-temporal Data (ICLR 2024 · similarity 0.90)
- Turbocharging Gaussian Process Inference with Approximate Sketch-and-Project (NeurIPS 2025 · similarity 0.87)
- Sparse Gaussian Processes: Structured Approximations and Power-EP Revisited (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
