add Simkus2025a and Kelly2025a

Author: michaelgutmann

_assets/NonRefereed.bib

@@ -48,22 +48,22 @@ @Inproceedings{Gutmann2009d
 }
 
 @Article{Gutmann2009c,
-  author        = {Gutmann, Michael and Hyv\"arinen, Aapo},
-  journal       = {BMC Neuroscience},
-  title         = {Learning spike-timings based representations of sensory stimuli with leaky integrate-and-fire neurons},
-  year          = {2009},
-  month         = jul,
-  number        = {1},
-  pages         = {P144},
-  volume        = {10},
-  __markedentry = {[mgutmann:]},
-  doi           = {https://doi.org/10.1186/1471-2202-10-S1-P144},
-  file          = {:/home/mgutmann/2-Reference/Michael-Gutmann/myPapers/Gutmann2009c.pdf:PDF},
-  issn          = {1471-2202},
-  owner         = {mgutmann},
-  refid         = {Gutmann2009},
-  timestamp     = {2020.09.14},
-  url           = {https://link.springer.com/article/10.1186/1471-2202-10-S1-P144},
+  author    = {Gutmann, Michael and Hyv\"arinen, Aapo},
+  journal   = {BMC Neuroscience},
+  title     = {Learning spike-timings based representations of sensory stimuli with leaky integrate-and-fire neurons},
+  year      = {2009},
+  issn      = {1471-2202},
+  month     = jul,
+  number    = {1},
+  pages     = {P144},
+  volume    = {10},
+  doi       = {https://doi.org/10.1186/1471-2202-10-S1-P144},
+  file      = {:/home/mgutmann/2-Reference/Michael-Gutmann/myPapers/Gutmann2009c.pdf:PDF},
+  groups    = {[mgutmann:]},
+  owner     = {mgutmann},
+  refid     = {Gutmann2009},
+  timestamp = {2020.09.14},
+  url       = {https://link.springer.com/article/10.1186/1471-2202-10-S1-P144},
 }
 
 @Article{Dinev2018,
@@ -104,4 +104,20 @@ @Article{Kleinegesse2021a
   url       = {https://arxiv.org/abs/2105.04379},
 }
 
+@Article{Simkus2025a,
+  author   = {Vaidotas Simkus and Michael U. Gutmann},
+  journal  = {arXiv:2506.09258},
+  title    = {CFMI: Flow Matching for Missing Data Imputation},
+  year     = {2025},
+  abstract = {We introduce conditional flow matching for imputation (CFMI), a new general-purpose method to impute missing data. The method combines continuous normalising flows, flow-matching, and shared conditional modelling to deal with intractabilities of traditional multiple imputation. Our comparison with nine classical and state-of-the-art imputation methods on 24 small to moderate-dimensional tabular data sets shows that CFMI matches or outperforms both traditional and modern techniques across a wide range of metrics. Applying the method to zero-shot imputation of time-series data, we find that it matches the accuracy of a related diffusion-based method while outperforming it in terms of computational efficiency. Overall, CFMI performs at least as well as traditional methods on lower-dimensional data while remaining scalable to high-dimensional settings, matching or exceeding the performance of other deep learning-based approaches, making it a go-to imputation method for a wide range of data types and dimensionalities.},
+  arxiv    = {https://arxiv.org/abs/2506.09258},
+  url      = {https://doi.org/10.48550/arXiv.2506.09258},
+}
+
 @Comment{jabref-meta: databaseType:bibtex;}
+
+@Comment{jabref-meta: grouping:
+0 AllEntriesGroup:;
+1 StaticGroup:Markings\;2\;1\;\;\;\;;
+2 StaticGroup:[mgutmann:]\;2\;1\;\;\;\;;
+}

_assets/UnderReview.bib

@@ -12,6 +12,16 @@ @Article{Kobiela2024a
   url      = {http://biorxiv.org/content/early/2024/11/13/2024.11.13.623219.abstract},
 }
 
+@Article{Kelly2025,
+  author   = {Ryan P. Kelly and David J. Warne and David T. Frazier and David J. Nott and Michael U. Gutmann and Christopher Drovandi},
+  journal  = {arXiv:2503.12315},
+  title    = {Simulation-based Bayesian inference under model misspecification},
+  year     = {2025},
+  abstract = {Simulation-based Bayesian inference (SBI) methods are widely used for parameter estimation in complex models where evaluating the likelihood is challenging but generating simulations is relatively straightforward. However, these methods commonly assume that the simulation model accurately reflects the true data-generating process, an assumption that is frequently violated in realistic scenarios. In this paper, we focus on the challenges faced by SBI methods under model misspecification. We consolidate recent research aimed at mitigating the effects of misspecification, highlighting three key strategies: i) robust summary statistics, ii) generalised Bayesian inference, and iii) error modelling and adjustment parameters. To illustrate both the vulnerabilities of popular SBI methods and the effectiveness of misspecification-robust alternatives, we present empirical results on an illustrative example.},
+  arxiv    = {https://arxiv.org/abs/2503.12315},
+  url      = {https://doi.org/10.48550/arXiv.2503.12315},
+}
+
 @Comment{jabref-meta: databaseType:bibtex;}
 
 @Comment{jabref-meta: fileDirectory-mgutmann-zermatt:../myPapers;}

index.md

@@ -19,8 +19,8 @@ A short biography is available [here](bio/index.html).
 
 ## Recent papers
 
+* arXiv: CFMI: Flow Matching for Missing Data Imputation [[link]](publications/index.html#Simkus2025a)
+* arXiv: Simulation-based Bayesian inference under model misspecification [[link]](publications/index.html#Kelly2025a)
 * BioRxiv: Risk-averse optimization of genetic circuits under uncertainty [[link]](publications/index.html#Kobiela2024a)
 * TMLR: Improving Variational Autoencoder Estimation from Incomplete Data with Mixture Variational Families [[link]](publications/index.html#Simkus2024a)
 * eLife: Designing Optimal Behavioral Experiments Using Machine Learning [[link]](publications/index.html#Valentin2023a)
-* TMLR: Conditional Sampling of Variational Autoencoders via Iterated Approximate Ancestral Sampling [[link]](publications/index.html#Simkus2023b)
-* JMLR: Variational Gibbs Inference for Statistical Model Estimation from Incomplete Data [[link]](publications/index.html#Simkus2023a)