5 papers
How to Approximate Inference with Subtractive Mixture Models
Lena Zellinger, Nicola Branchini, Lennert De Smet +3
Classical mixture models (MMs) are widely used tractable proposals for approximate inference settings such as variational inference (VI) and importance sampling (IS). Recently, mix…
Multi-Marginal Flow Matching with Adversarially Learnt Interpolants
Oskar Kviman, Kirill Tamogashev, Nicola Branchini +3
Learning the dynamics of a process given sampled observations at several time points is an important but difficult task in many scientific applications. When no ground-truth trajec…
Towards Adaptive Self-Normalized Importance Samplers
Nicola Branchini, VÃctor Elvira
The self-normalized importance sampling (SNIS) estimator is a Monte Carlo estimator widely used to approximate expectations in statistical signal processing and machine learning. T…
Scalable Expectation Estimation with Subtractive Mixture Models
Lena Zellinger, Nicola Branchini, VÃctor Elvira +1
Many Monte Carlo (MC) and importance sampling (IS) methods use mixture models (MMs) for their simplicity and ability to capture multimodal distributions. Recently, subtractive mixt…
Generalizing self-normalized importance sampling with couplings
Nicola Branchini, VÃctor Elvira
An essential problem in statistics and machine learning is the estimation of expectations involving PDFs with intractable normalizing constants. The self-normalized importance samp…