OxCSML : People : Sarah Filippi

Statistical machine learning and Bayesian statistics motivated by applications in biomedicine

I am a Lecturer in Biostatistics at Imperial College London. I hold a joint position between the Statistic Section of the Department of Mathematics and the Department of Epidemiology and Biostatistics. The core of my research lies in statistical machine learning and computational statistics methodology motivated by applications in and around computational biology and biomedical genetics. I am particularly interested in addressing how novel statistical and computational approaches and algorithms can aid in the analysis of large-scale real-world biomedical data.

Publications

2018

Q. Zhang , S. Filippi , A. Gretton , D. Sejdinovic , Large-Scale Kernel Methods for Independence Testing, Statistics and Computing, vol. 28, no. 1, 113–130, Jan. 2018.
Project: bigbayes

Representations of probability measures in reproducing kernel Hilbert spaces provide a flexible framework for fully nonparametric hypothesis tests of independence, which can capture any type of departure from independence, including nonlinear associations and multivariate interactions. However, these approaches come with an at least quadratic computational cost in the number of observations, which can be prohibitive in many applications. Arguably, it is exactly in such large-scale datasets that capturing any type of dependence is of interest, so striking a favourable tradeoff between computational efficiency and test performance for kernel independence tests would have a direct impact on their applicability in practice. In this contribution, we provide an extensive study of the use of large-scale kernel approximations in the context of independence testing, contrasting block-based, Nystrom and random Fourier feature approaches. Through a variety of synthetic data experiments, it is demonstrated that our novel large scale methods give comparable performance with existing methods whilst using significantly less computation time and memory.
```
@article{ZhaFilGreSej2018,
  author = {Zhang, Q. and Filippi, S. and Gretton, A. and Sejdinovic, D.},
  journal = {Statistics and Computing},
  year = {2018},
  month = jan,
  volume = {28},
  number = {1},
  pages = {113--130},
  title = {{Large-Scale Kernel Methods for Independence Testing}}
}
```

2017

Q. F. Wills , E. Mellado-Gomez , R. Nolan , D. Warner , E. Sharma , J. Broxholme , B. Wright , H. Lockstone , W. James , M. Lynch , M. Gonzales , J. West , A. Leyrat , S. Padilla-Parra , S. Filippi , C. Holmes , M. D. Moore , R. Bowden , The nature and nurture of cell heterogeneity: accounting for macrophage gene-environment interactions with single-cell RNA-Seq, BMC Genomics, 2017.

@article{Wills2017,
  author = {Wills, Quin F. and Mellado-Gomez, Esther and Nolan, Rory and Warner, Damien and Sharma, Eshita and Broxholme, John and Wright, Benjamin and Lockstone, Helen and James, William and Lynch, Mark and Gonzales, Michael and West, Jay and Leyrat, Anne and Padilla-Parra, Sergi and Filippi, Sarah and Holmes, Chris and Moore, Michael D. and Bowden, Rory},
  title = {The nature and nurture of cell heterogeneity: accounting for macrophage gene-environment interactions with single-cell RNA-Seq},
  journal = {BMC Genomics},
  year = {2017}
}

S. Filippi , C. C. Holmes , . others , A Bayesian nonparametric approach to testing for dependence between random variables, Bayesian Analysis, 2017.

@article{filippi2017bayesian,
  title = {A Bayesian nonparametric approach to testing for dependence between random variables},
  author = {Filippi, Sarah and Holmes, Chris C and others},
  journal = {Bayesian Analysis},
  year = {2017},
  publisher = {International Society for Bayesian Analysis}
}

Q. Zhang , S. Filippi , S. Flaxman , D. Sejdinovic , Feature-to-Feature Regression for a Two-Step Conditional Independence Test, in Uncertainty in Artificial Intelligence (UAI), 2017.
Project: bigbayes

The algorithms for causal discovery and more broadly for learning the structure of graphical models require well calibrated and consistent conditional independence (CI) tests. We revisit the CI tests which are based on two-step procedures and involve regression with subsequent (unconditional) independence test (RESIT) on regression residuals and investigate the assumptions under which these tests operate. In particular, we demonstrate that when going beyond simple functional relationships with additive noise, such tests can lead to an inflated number of false discoveries. We study the relationship of these tests with those based on dependence measures using reproducing kernel Hilbert spaces (RKHS) and propose an extension of RESIT which uses RKHS-valued regression. The resulting test inherits the simple two-step testing procedure of RESIT, while giving correct Type I control and competitive power. When used as a component of the PC algorithm, the proposed test is more robust to the case where hidden variables induce a switching behaviour in the associations present in the data.
```
@inproceedings{ZhaFilFlaSej2017,
  author = {Zhang, Q. and Filippi, S. and Flaxman, S. and Sejdinovic, D.},
  title = {{Feature-to-Feature Regression for a Two-Step Conditional Independence Test}},
  booktitle = {Uncertainty in Artificial Intelligence (UAI)},
  year = {2017}
}
```

2016

S. Filippi , C. P. Barnes , P. D. W. Kirk , T. Kudo , K. Kunida , S. S. McMahon , T. Tsuchiya , T. Wada , S. Kuroda , M. P. H. Stumpf , Robustness of MEK-ERK Dynamics and Origins of Cell-to-Cell Variability in MAPK Signaling, CellReports, 2016.

@article{Filippi:2016gs,
  author = {Filippi, Sarah and Barnes, Chris P and Kirk, Paul D W and Kudo, Takamasa and Kunida, Katsuyuki and McMahon, Siobhan S and Tsuchiya, Takaho and Wada, Takumi and Kuroda, Shinya and Stumpf, Michael P H},
  title = {{Robustness of MEK-ERK Dynamics and Origins of Cell-to-Cell Variability in MAPK Signaling}},
  journal = {CellReports},
  year = {2016}
}

S. Filippi , C. C. Holmes , L. E. Nieto-Barajas , Scalable Bayesian nonparametric measures for exploring pairwise dependence via Dirichlet Process Mixtures, Electronic Journal of Statistics, 2016.

@article{Filippi:2016kg,
  author = {Filippi, Sarah and Holmes, Chris C and Nieto-Barajas, Luis E.},
  title = {{Scalable Bayesian nonparametric measures for exploring pairwise dependence via Dirichlet Process Mixtures}},
  journal = {Electronic Journal of Statistics},
  year = {2016}
}

S. Filippi , C. C. Holmes , A Bayesian Nonparametric Approach to Testing for Dependence Between Random Variables, Bayesian Analysis, 2016.

@article{Filippi:2016di,
  author = {Filippi, Sarah and Holmes, Chris C},
  title = {{A Bayesian Nonparametric Approach to Testing for Dependence Between Random Variables}},
  journal = {Bayesian Analysis},
  year = {2016}
}

S. Filippi , C. C. Holmes , L. E. Nieto-Barajas , . others , Scalable Bayesian nonparametric measures for exploring pairwise dependence via Dirichlet Process Mixtures, Electronic Journal of Statistics, vol. 10, no. 2, 3338–3354, 2016.

@article{filippi2016scalable,
  title = {Scalable Bayesian nonparametric measures for exploring pairwise dependence via Dirichlet Process Mixtures},
  author = {Filippi, Sarah and Holmes, Chris C and Nieto-Barajas, Luis E and others},
  journal = {Electronic Journal of Statistics},
  volume = {10},
  number = {2},
  pages = {3338--3354},
  year = {2016},
  publisher = {The Institute of Mathematical Statistics and the Bernoulli Society}
}

S. Flaxman , D. Sejdinovic , J. Cunningham , S. Filippi , Bayesian Learning of Kernel Embeddings, in Uncertainty in Artificial Intelligence (UAI), 2016, 182–191.
Project: bigbayes

Kernel methods are one of the mainstays of machine learning, but the problem of kernel learning remains challenging, with only a few heuristics and very little theory. This is of particular importance in methods based on estimation of kernel mean embeddings of probability measures. For characteristic kernels, which include most commonly used ones, the kernel mean embedding uniquely determines its probability measure, so it can be used to design a powerful statistical testing framework, which includes nonparametric two-sample and independence tests. In practice, however, the performance of these tests can be very sensitive to the choice of kernel and its lengthscale parameters. To address this central issue, we propose a new probabilistic model for kernel mean embeddings, the Bayesian Kernel Embedding model, combining a Gaussian process prior over the Reproducing Kernel Hilbert Space containing the mean embedding with a conjugate likelihood function, thus yielding a closed form posterior over the mean embedding. The posterior mean of our model is closely related to recently proposed shrinkage estimators for kernel mean embeddings, while the posterior uncertainty is a new, interesting feature with various possible applications. Critically for the purposes of kernel learning, our model gives a simple, closed form marginal pseudolikelihood of the observed data given the kernel hyperparameters. This marginal pseudolikelihood can either be optimized to inform the hyperparameter choice or fully Bayesian inference can be used.
```
@inproceedings{FlaSejCunFil2016,
  author = {Flaxman, S. and Sejdinovic, D. and Cunningham, J.P. and Filippi, S.},
  booktitle = {Uncertainty in Artificial Intelligence (UAI)},
  pages = {182--191},
  title = {{Bayesian Learning of Kernel Embeddings}},
  year = {2016},
  bdsk-url-1 = {http://www.auai.org/uai2016/proceedings/papers/145.pdf},
  bdsk-url-2 = {http://www.auai.org/uai2016/proceedings/supp/145_supp.pdf}
}
```

2015

S. S. M. Mahon , O. Lenive , S. Filippi , M. P. H. Stumpf , Information processing by simple molecular motifs and susceptibility to noise, Journal of The Royal Society Interface, 2015.

@article{Mahon:2015hz,
  author = {Mahon, Siobhan S Mc and Lenive, Oleg and Filippi, Sarah and Stumpf, Michael P H},
  title = {{Information processing by simple molecular motifs and susceptibility to noise}},
  journal = {Journal of The Royal Society Interface},
  year = {2015}
}

2014

S. S. Mc Mahon , A. Sim , S. Filippi , R. Johnson , J. Liepe , D. Smith , M. P. H. Stumpf , Information theory and signal transduction systems: From molecular information processing to network inference., Seminars in cell & developmental biology, 2014.

@article{McMahon:2014cs,
  author = {Mc Mahon, Siobhan S and Sim, Aaron and Filippi, Sarah and Johnson, Robert and Liepe, Juliane and Smith, Dominic and Stumpf, Michael P H},
  title = {{Information theory and signal transduction systems: From molecular information processing to network inference.}},
  journal = {Seminars in cell {\&} developmental biology},
  year = {2014}
}

A. L. MacLean , S. Filippi , M. P. H. Stumpf , The ecology in the hematopoietic stem cell niche determines the clinical outcome in chronic myeloid leukemia, PNAS, 2014.

@article{MacLean:2014if,
  author = {MacLean, Adam L and Filippi, Sarah and Stumpf, Michael P H},
  title = {{The ecology in the hematopoietic stem cell niche determines the clinical outcome in chronic myeloid leukemia}},
  journal = {PNAS},
  year = {2014}
}

J. Liepe , P. Kirk , S. Filippi , T. Toni , C. P. Barnes , M. P. H. Stumpf , A framework for parameter estimation and model selection from experimental data in systems biology using approximate Bayesian computation., Nature Protocols, 2014.

@article{Liepe:2014iw,
  author = {Liepe, Juliane and Kirk, Paul and Filippi, Sarah and Toni, Tina and Barnes, Chris P and Stumpf, Michael P H},
  title = {{A framework for parameter estimation and model selection from experimental data in systems biology using approximate Bayesian computation.}},
  journal = {Nature Protocols},
  year = {2014}
}

2013

D. Silk , S. Filippi , M. P. H. Stumpf , Optimizing threshold-schedules for sequential approximate Bayesian computation: applications to molecular systems., Statistical Applications in Genetics and Molecular Biology, 2013.

@article{Silk:2013bi,
  author = {Silk, Daniel and Filippi, Sarah and Stumpf, Michael P H},
  title = {{Optimizing threshold-schedules for sequential approximate Bayesian computation: applications to molecular systems.}},
  journal = {Statistical Applications in Genetics and Molecular Biology},
  year = {2013}
}

S. Filippi , C. P. Barnes , J. Cornebise , M. P. H. Stumpf , On optimality of kernels for approximate Bayesian computation using sequential Monte Carlo., Statistical Applications in Genetics and Molecular Biology, 2013.

@article{Filippi:2013io,
  author = {Filippi, Sarah and Barnes, Chris P and Cornebise, Julien and Stumpf, Michael P H},
  title = {{On optimality of kernels for approximate Bayesian computation using sequential Monte Carlo.}},
  journal = {Statistical Applications in Genetics and Molecular Biology},
  year = {2013}
}

J. Liepe , S. Filippi , M. Komorowski , M. P. H. Stumpf , Maximizing the Information Content of Experiments in Systems Biology, PLoS computational biology, 2013.

@article{Liepe:2013cl,
  author = {Liepe, Juliane and Filippi, Sarah and Komorowski, Micha{\l} and Stumpf, Michael P H},
  title = {{Maximizing the Information Content of Experiments in Systems Biology}},
  journal = {PLoS computational biology},
  year = {2013}
}

2012

A. Roy , G. Cowan , A. J. Mead , S. Filippi , G. Bohn , A. Chaidos , O. Tunstall , J. K. Y. Chan , M. Choolani , P. Bennett , S. Kumar , D. Atkinson , J. Wyatt-Ashmead , M. Hu , M. P. H. Stumpf , K. Goudevenou , D. O’Connor , S. T. Chou , M. J. Weiss , A. Karadimitris , S. E. Jacobsen , P. Vyas , I. Roberts , Perturbation of fetal liver hematopoietic stem and progenitor cell development by trisomy 21., Proceedings of the National Academy of Sciences, 2012.

@article{Roy:2012fs,
  author = {Roy, Anindita and Cowan, Gillian and Mead, Adam J and Filippi, Sarah and Bohn, Georg and Chaidos, Aristeidis and Tunstall, Oliver and Chan, Jerry K Y and Choolani, Mahesh and Bennett, Phillip and Kumar, Sailesh and Atkinson, Deborah and Wyatt-Ashmead, Josephine and Hu, Ming and Stumpf, Michael P H and Goudevenou, Katerina and O'Connor, David and Chou, Stella T and Weiss, Mitchell J and Karadimitris, Anastasios and Jacobsen, Sten Eirik and Vyas, Paresh and Roberts, Irene},
  title = {{Perturbation of fetal liver hematopoietic stem and progenitor cell development by trisomy 21.}},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2012}
}

C. P. Barnes , S. Filippi , M. P. H. Stumpf , T. Thorne , Considerate approaches to constructing summary statistics for ABC model selection, Statistics and Computing, 2012.

@article{Barnes:2012hf,
  author = {Barnes, Chris P and Filippi, Sarah and Stumpf, Michael P H and Thorne, Thomas},
  title = {{Considerate approaches to constructing summary statistics for ABC model selection}},
  journal = {Statistics and Computing},
  year = {2012}
}

2011

S. Filippi , O. Cappe , A. Garivier , Optimally Sensing a Single Channel Without Prior Information: The Tiling Algorithm and Regret Bounds, IEEE Journal of Selected Topics in Signal Processing, 2011.

@article{Filippi:2011bd,
  author = {Filippi, Sarah and Cappe, Olivier and Garivier, Aurelien},
  title = {{Optimally Sensing a Single Channel Without Prior Information: The Tiling Algorithm and Regret Bounds}},
  journal = {IEEE Journal of Selected Topics in Signal Processing},
  year = {2011}
}

2010

S. Filippi , O. Cappe , A. Garivier , C. Szepesvari , Parametric bandits: The generalized linear case, in Neural Information Processing Systems (NIPS’2010), 2010.

@inproceedings{Filippi:2010un,
  author = {Filippi, S and Cappe, O and Garivier, A and Szepesvari, C},
  title = {{Parametric bandits: The generalized linear case}},
  booktitle = {Neural Information Processing Systems (NIPS'2010)},
  year = {2010}
}

S. Filippi , O. Cappe , A. Garivier , Optimism in reinforcement learning and Kullback-Leibler divergence, in 2010 48th Annual Allerton Conference on Communication, Control, and Computing (Allerton), 2010.

@inproceedings{Filippi:2010gj,
  author = {Filippi, Sarah and Cappe, Olivier and Garivier, Aurelien},
  title = {{Optimism in reinforcement learning and Kullback-Leibler divergence}},
  booktitle = {2010 48th Annual Allerton Conference on Communication, Control, and Computing (Allerton)},
  year = {2010}
}

2008

S. Filippi , O. Cappe , F. Clerot , E. Moulines , A Near Optimal Policy for Channel Allocation in Cognitive Radio, in Lecture Notes in Computer Science, Recent Advances in Reinforcement Learning, Springer, 2008.

@inproceedings{Filippi:2008em,
  author = {Filippi, Sarah and Cappe, Olivier and Clerot, Fabrice and Moulines, Eric},
  title = {{A Near Optimal Policy for Channel Allocation in Cognitive Radio}},
  booktitle = {Lecture Notes in Computer Science, Recent Advances in Reinforcement Learning, Springer},
  year = {2008}
}