Computation

New submissions

• New submissions
• Replacements

New submissions for Fri, 19 Apr 24

[1]  arXiv:2404.12290 [pdf, other]

Title: Debiased Distribution Compression

Authors: Lingxiao Li, Raaz Dwivedi, Lester Mackey

Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Computation (stat.CO); Methodology (stat.ME)

Modern compression methods can summarize a target distribution $\mathbb{P}$ more succinctly than i.i.d. sampling but require access to a low-bias input sequence like a Markov chain converging quickly to $\mathbb{P}$. We introduce a new suite of compression methods suitable for compression with biased input sequences. Given $n$ points targeting the wrong distribution and quadratic time, Stein Kernel Thinning (SKT) returns $\sqrt{n}$ equal-weighted points with $\widetilde{O}(n^{-1/2})$ maximum mean discrepancy (MMD) to $\mathbb {P}$. For larger-scale compression tasks, Low-rank SKT achieves the same feat in sub-quadratic time using an adaptive low-rank debiasing procedure that may be of independent interest. For downstream tasks that support simplex or constant-preserving weights, Stein Recombination and Stein Cholesky achieve even greater parsimony, matching the guarantees of SKT with as few as $\operatorname{poly-log}(n)$ weighted points. Underlying these advances are new guarantees for the quality of simplex-weighted coresets, the spectral decay of kernel matrices, and the covering numbers of Stein kernel Hilbert spaces. In our experiments, our techniques provide succinct and accurate posterior summaries while overcoming biases due to burn-in, approximate Markov chain Monte Carlo, and tempering.

Replacements for Fri, 19 Apr 24

[2]  arXiv:2306.00389 (replaced) [pdf, other]

Title: R-VGAL: A Sequential Variational Bayes Algorithm for Generalised Linear Mixed Models

Authors: Bao Anh Vu, David Gunawan, Andrew Zammit-Mangion

Subjects: Computation (stat.CO); Methodology (stat.ME)

[3]  arXiv:2402.14390 (replaced) [pdf, other]

Title: Composite likelihood inference for the Poisson log-normal model

Authors: Julien Stoehr (CEREMADE), Stephane S. Robin (LPSM (UMR\_8001))

Subjects: Computation (stat.CO); Methodology (stat.ME)

[4]  arXiv:2402.06885 (replaced) [pdf, other]

Title: DimVis: Interpreting Visual Clusters in Dimensionality Reduction With Explainable Boosting Machine

Authors: Parisa Salmanian, Angelos Chatzimparmpas, Ali Can Karaca, Rafael M. Martins

Comments: This manuscript is accepted for publication in EuroVis 2024 MLVis Workshop

Subjects: Human-Computer Interaction (cs.HC); Machine Learning (cs.LG); Computation (stat.CO)

• New submissions
• Replacements