Amortized Variational Inference of Latent Intensities from Replicated Point Patterns
We propose a neural-network-based amortized variational framework that learns a reusable inference rule mapping a variable number of replicated point patterns to a distribution over their shared latent intensity.
ProofPaper ↗
Key points
- Bayesian inference for latent point process intensities often relies on per-dataset sampling or optimization, making repeated inference computationally expensive.
- Trained from event times using the point process likelihood, the model learns an amortized inference rule without ground-truth intensity supervision.
- At test time, it performs inference of the latent distribution in a forward pass.
- On synthetic data, additional realizations improve intensity recovery and yield narrower credible intervals with broadly stable coverage.
Sources (1)
- [1]Amortized Variational Inference of Latent Intensities from Replicated Point PatternsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 12, 12:18 PM
We propose a neural-network-based amortized variational framework that learns a reusable inference rule mapping a variable number of replicated point patterns to a distribution over their shared latent intensity.
Bayesian inference for latent point process intensities often relies on per-dataset sampling or optimization, making repeated inference computationally expensive.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 11, 2026Teaching Agents to Search with NVIDIA Data Designer — Dhruv Nathawani, NVIDIA
- Oct 11, 2026The Mathematical Collective Unconscious: Jungian Archetypes within the Latent Space
- Oct 11, 2026Screening Collapse under Recursive Synthetic Data
- Oct 10, 2026Vela: User Behavior as Supervision Sparks LLM Intelligence
- Oct 9, 2026Microsoft's Decision-1 model enters the fast-growing AI decision model race
- Oct 9, 2026X-AVDD: Cross-Attentive Audio-Visual Dataset Distillation