Publications
This page contains a list of my publications in a chronological order. Click here to see my unpublished works. If you have questions regarding any of these publications, please feel free to contact me.
2019

Witnessing Adversarial Training in Reproducing Kernel Hilbert Spaces
A. Mehrjou, W. Jitkrittum, B. Schölkopf, K. Muandet
Under review
eprint

Improving Consequential Decision Making under Imperfect Predictions
N. Kilbertus, M. Gomez Rodriguez, B Schölkopf, K. Muandet, I. Valera
Under review

Local Temporal Bilinear Pooling for Finegrained Action Parsing
Y. Zhang, S. Tang, K. Muandet, C. Jarvers, and H. Neumann
IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR2019)
eprint
2018

Counterfactual Mean Embedding: A Kernel Method for Nonparametric Causal Inference
K. Muandet, M. Kanagawa, S. Saengkyongam, and S. Marukatat
Under review
eprint

Design and Analysis of the NIPS 2016 Review Process
N. Shah, B. Tabibian, K. Muandet, I. Guyon, and U. von Luxburg
Journal of Machine Learning Research, 19(49):1−34, 2018.
eprint
link
2017

Eigendecompositions of Transfer Operators in Reproducing Kernel Hilbert Spaces
S. Klus, I. Schuster, and K. Muandet
Submitted
eprint

Minimax Estimation of Kernel Mean Embeddings
I. Tolstikhin, B. Sriperumbudur, and K. Muandet
Journal of Machine Learning Research, 18(86):1−47, 2017.
eprint

Kernel Mean Embedding of Distributions : A Review and Beyond
K. Muandet, K. Fukumizu, B. Sriperumbudur, and B. Schölkopf
Foundations and Trends in Machine Learning, Vol. 10: No. 12, pp 1141.
eprint
link

Dagstuhl Seminar: New Directions for Learning with Kernels and Gaussian Processes
A. Gretton, P. Hennig, C. E. Rasmussen, and B. Schölkopf
Report of Dagstuhl Seminar 16481 (contributed talk on kernel mean shrinkage estimators)
eprint
link
2016

A Scalable Mixednorm Approach for Learning Lightweight Models in Largescale Classification
R. Babber, K. Muandet, and B. Schölkopf
SIAM International Conference on Data Mining (SDM2016)
eprint
2015

From Points to Probability Measures: Statistical Learning on Distributions with Kernel Mean Embedding
K. Muandet
Ph.D. Thesis. Department of Computer Science, University of Tübingen
eprint
PDF
 Kernel Mean Shrinkage Estimators
K. Muandet*, B. Sriperumbudur*, K. Fukumizu, A. Gretton, and B. Schölkopf (* contributed equally)
Journal of Machine Learning Research
eprint

Towards a Learning Theory of CauseEffect Inference
D. LopezPaz, K. Muandet, B. Schölkopf, and Iliya Tolstikhin
Proceeding of the 32nd International Conference on Machine Learning (ICML 2015)
abstract
eprint
bib
slide
poster
We pose causal inference as the problem of learning to classify probability distributions. In particular,
we assume access to a collection $\{(S_i, l_i)\}^n_{i=1}$, where each $S_i$ is a sample drawn from the
probability distribution of $X_i\times Y_i$, and $l_i$ is a binary label indicating whether
"$X_i \rightarrow Y_i$" or "$X_i \leftarrow Y_i$". Given these data, we build a causal inference rule in
two steps. First, we featurize each Si using the kernel mean embedding associated with some characteristic
kernel. Second, we train a binary classifier on such embeddings to distinguish between causal directions.
We present generalization bounds showing the statistical consistency and learning rates of the proposed
approach, and provide a simple implementation that achieves stateoftheart causeeffect inference.
Furthermore, we extend our ideas to infer causal relationships between more than two variables.

Computing Functions of Random Variables via Reproducing Kernel Hilbert Space Representations
B. Schölkopf, K. Muandet, K. Fukumizu, and J. Peters
In Statistics and Computing
abstract
eprint
bib
We describe a method to perform functional operations on probability distributions of random variables. The method uses reproducing kernel Hilbert space representations of probability distributions, and it is applicable to all operations which can be applied to points drawn from the respective distributions. We refer to our approach as kernel probabilistic programming. We illustrate it on synthetic data, and show how it can be used for nonparametric structural equation models, with an application to causal inference.
2014
 SingleSource Domain Adaptation with Target and Conditional Shift
K. Zhang, B. Schölkopf, K. Muandet, Z. Wang, Z. Zhou, and C. Persello
In Regularization, Optimization, Kernels, and Support Vector Machines
(Ed) JAK Suykens, M Signoretto and A Argyriou, Chapman and Hall/CRC, Boca Raton, USA, (2014)
[pdf]
[bib]
 The Randomized Causation Coefficient
D. LopezPaz, K. Muandet, and B. Recht
Journal of Machine Learning Research (Special Topic on Causality)
eprint
PDF
bib
 Kernel Mean Estimation via Spectral Filtering
K. Muandet, B. Sriperumbudur, and B. Schölkopf
In the 28th Annual Conference on Neural Information Processing Systems (NIPS 2014)
[pdf]
[supplementary]
 A Permutationbased Kernel Conditional Independence Test
G. Doran, K. Muandet, K. Zhang, and B. Schölkopf
In Proceeding of the 30th Conference on Uncertainty in Artificial Intelligence (UAI 2014)
[pdf]
[supplementary]
[bib]
[code]
 Kernel Mean Estimation and
Stein Effect
K. Muandet, K. Fukumizu, B. Sriperumbudur, A. Gretton, and B. Schölkopf
In Proceeding of the 31st International Conference on Machine Learning (ICML 2014)
[pdf]
[supplementary]
[bib]
[code]
2013
 Domain Adaptation under Target and Conditional Shift
K. Zhang, B. Schölkopf, K. Muandet, and Z. Wang
In Proceeding of the 30th International Conference on Machine Learning (ICML 2013)
[pdf]
[supplementary]
[bib]
 Oneclass Support Measure Machines for Group Anomaly Detection
K. Muandet and B. Schölkopf
In Proceeding of the 29th Conference on Uncertainty in Artificial Intelligence (UAI 2013)
[pdf]
[bib]
 Domain Generalization via Invariant Feature Representation
K. Muandet, D. Balduzzi, and B. Schölkopf
In Proceeding of the 30th International Conference on Machine Learning (ICML 2013)
[pdf]
[supplementary]
[bib]
[slide]
[poster]
[code]
2012
 Learning from Distributions via Support Measure Machines
K. Muandet, K. Fukumizu, F. Dinuzzo, and B. Schölkopf
In Proceeding of the 26th Annual Conference on Neural Information Processing Systems (NIPS 2012)
[pdf]
[supplementary]
[bib]
[spotlight]
[poster]
[code]
 Hilbert Space Embedding for Dirichlet Process Mixtures
K. Muandet
In NIPS2012 Workshop on Confluence between Kernel Methods and Graphical Models
[pdf]
[bib]
[code]
2010
 Infinite Independent Subspace Analysis
K. Muandet and Y.W.Teh
M.Sc. Thesis, Department of Computer Science, University College London, 2010.
[pdf]
[bib]
2009
 Query Selection via Weighted Entropy for GraphBased Semisupervised Classification
K. Muandet, S. Marukatat, and C. Nattee
In Proceedings of Asian Conference on Machine Learning (ACML), pages 278292. Springer Press, 2009.
[pdf]
[bib]
 Robust Graph Hyperparameter Learning for Graph Based Semisupervised Classification
K.Muandet, S.Marukatat, and C.Nattee
In Proceedings of PacificAsia Conference on Knowledge Discovery and Data Mining conference (PAKDD),
pages 278292. Springer Press, 2009.
[pdf]
[bib]
2008
 PACS (Picture Archiving Communication System) for dentistry
N. Patanachai, B. Uyyanonvara, C. Sinthanayothin, W. Tharanon, P. Sompot, and K. Muandet
[pdf]
[bib]
2007
 Development of Dental Software: Introducing ADTEC DICOM Viewer
C. Sinthanayothin, K. Muandet, B. Uyyanonvara, and W. Tharanon
[pdf]
[bib]
My bibliographic information is also available on DBLP.
Unpublished Works

A Unifying View of Support Measure Machines, Support Vector Machines, and Parzen Window Classifiers
K. Muandet and B. Schölkopf
abstract
pdf
bib
This paper presents a unifying view of two wellknown kernelbased classifiers,
namely support vector machines (SVMs) and Parzen window classifiers. In particular, given
the training data, both learning algorithms can be viewed as a solution to a regularization
problem on probability distributions, depending on how the distributions are constructed from
the training data. This simple insight may shed light on the unification of various kernelbased
learning algorithms.