no code implementations • ICML 2020 • Tim G. J. Rudner, Dino Sejdinovic, Yarin Gal
We propose Inter-domain Deep Gaussian Processes with RKHS Fourier Features, an extension of shallow inter-domain GPs that combines the advantages of inter-domain and deep Gaussian processes (DGPs) and demonstrate how to leverage existing approximate inference approaches to perform simple and scalable approximate inference on Inter-domain Deep Gaussian Processes.
no code implementations • 30 Oct 2024 • Dino Sejdinovic
Kernel embeddings have emerged as a powerful tool for representing probability measures in a variety of statistical inference problems.
1 code implementation • 16 Oct 2024 • Siu Lun Chau, Antonin Schrab, Arthur Gretton, Dino Sejdinovic, Krikamol Muandet
We introduce credal two-sample testing, a new hypothesis testing framework for comparing credal sets -- convex sets of probability measures where each element captures aleatoric uncertainty and the set itself represents epistemic uncertainty that arises from the modeller's partial ignorance.
no code implementations • 30 Jul 2024 • Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo, Arash Mohammadi, Hamid Alinejad-Rokny, Dino Sejdinovic, Damith C. Ranasinghe, Ehsan Abbasnejad
Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks.
no code implementations • 23 May 2024 • Rafael Oliveira, Dino Sejdinovic, David Howard, Edwin Bonilla
The process of calibrating computer models of natural phenomena is essential for applications in the physical sciences, where plenty of domain knowledge can be embedded into simulations and then calibrated against real observations.
no code implementations • 16 Mar 2024 • Eiki Shimizu, Kenji Fukumizu, Dino Sejdinovic
In conditional density estimation tasks, our NN-CME hybrid achieves competitive performance and often surpasses existing deep learning-based methods.
1 code implementation • 14 Feb 2024 • Russell Tsuchida, Cheng Soon Ong, Dino Sejdinovic
Maximum likelihood and maximum a posteriori estimates in a reparameterisation of the final layer of the intensity function can be obtained by solving a (strongly) convex optimisation problem using projected gradient descent.
1 code implementation • 14 Jul 2023 • Shahine Bouabid, Dino Sejdinovic, Duncan Watson-Parris
The result is an emulator that \textit{(i)} enjoys the flexibility of statistical machine learning models and can learn from data, and \textit{(ii)} has a robust physical grounding with interpretable parameters that can be used to make inference about the climate system.
no code implementations • NeurIPS 2023 • Veit David Wild, Sahra Ghalebikesabi, Dino Sejdinovic, Jeremias Knoblauch
We establish the first mathematically rigorous link between Bayesian, variational Bayesian, and ensemble methods.
1 code implementation • 26 Jan 2023 • Shahine Bouabid, Jake Fawkes, Dino Sejdinovic
A directed acyclic graph (DAG) provides valuable prior knowledge that is often discarded in regression tasks in machine learning.
1 code implementation • 9 Dec 2022 • Jake Fawkes, Robert Hu, Robin J. Evans, Dino Sejdinovic
These improved estimators are inspired by doubly robust estimators of the causal mean, using a similar form within the kernel space.
no code implementations • 2 Nov 2022 • Diego Martinez-Taboada, Dino Sejdinovic
The framework naturally extends to the setting where one observes multiple treatment effects (e. g. an intermediate effect after an interim period, and an ultimate treatment effect which is of main interest) and allows for additionally modelling uncertainty about the relationship of these effects.
no code implementations • 25 Oct 2022 • Diego Martinez-Taboada, Dino Sejdinovic
The problem of sequentially maximizing the expectation of a function seeks to maximize the expected value of a function of interest without having direct control on its features.
no code implementations • 7 Aug 2022 • Marcos Matabuena, J. C Vidal, Oscar Hernan Madrid Padilla, Dino Sejdinovic
Biclustering algorithms partition data and covariates simultaneously, providing new insights in several domains, such as analyzing gene expression to discover new biological functions.
no code implementations • 22 Jun 2022 • Antonin Schrab, Wittawat Jitkrittum, Zoltán Szabó, Dino Sejdinovic, Arthur Gretton
We discuss how MultiFIT, the Multiscale Fisher's Independence Test for Multivariate Dependence proposed by Gorsky and Ma (2022), compares to existing linear-time kernel tests based on the Hilbert-Schmidt independence criterion (HSIC).
1 code implementation • 26 May 2022 • Robert Hu, Siu Lun Chau, Jaime Ferrando Huertas, Dino Sejdinovic
While preference modelling is becoming one of the pillars of machine learning, the problem of preference explanation remains challenging and underexplored.
1 code implementation • 12 May 2022 • Veit D. Wild, Robert Hu, Dino Sejdinovic
We develop a framework for generalized variational inference in infinite-dimensional function spaces and use it to construct a method termed Gaussian Wasserstein inference (GWI).
1 code implementation • 6 May 2022 • Shahine Bouabid, Duncan Watson-Parris, Sofija Stefanović, Athanasios Nenes, Dino Sejdinovic
In this work, we develop a framework for the vertical disaggregation of AOD into extinction profiles, i. e. the measure of light extinction throughout an atmospheric column, using readily available vertically resolved meteorological predictors such as temperature, pressure or relative humidity.
no code implementations • 28 Feb 2022 • Jake Fawkes, Robin Evans, Dino Sejdinovic
In order to achieve this, we must have causal models that are able to capture what someone would be like if we were to counterfactually change these traits.
no code implementations • 2 Feb 2022 • Robert Hu, Siu Lun Chau, Dino Sejdinovic, Joan Alexis Glaunès
Kernel matrix-vector multiplication (KMVM) is a foundational operation in machine learning and scientific computing.
1 code implementation • 1 Feb 2022 • Jonas Schuff, Dominic T. Lennon, Simon Geyer, David L. Craig, Federico Fedele, Florian Vigneau, Leon C. Camenzind, Andreas V. Kuhlmann, G. Andrew D. Briggs, Dominik M. Zumbühl, Dino Sejdinovic, Natalia Ares
Pauli spin blockade (PSB) can be employed as a great resource for spin qubit initialisation and readout even at elevated temperatures but it can be difficult to identify.
1 code implementation • 25 Nov 2021 • Robert Hu, Dino Sejdinovic, Robin J. Evans
Causal inference grows increasingly complex as the number of confounders increases.
no code implementations • 18 Oct 2021 • Siu Lun Chau, Robert Hu, Javier Gonzalez, Dino Sejdinovic
Feature attribution for kernel methods is often heuristic and not individualised for each prediction.
no code implementations • NeurIPS 2021 • Siu Lun Chau, Jean-François Ton, Javier González, Yee Whye Teh, Dino Sejdinovic
While causal models are becoming one of the mainstays of machine learning, the problem of uncertainty quantification in causal inference remains challenging.
no code implementations • 2 Jun 2021 • Veit Wild, Motonobu Kanagawa, Dino Sejdinovic
We investigate the connections between sparse approximation methods for making kernel methods and Gaussian processes (GPs) scalable to large-scale data, focusing on the Nystr\"om method and the Sparse Variational Gaussian Processes (SVGP).
1 code implementation • NeurIPS 2021 • Siu Lun Chau, Shahine Bouabid, Dino Sejdinovic
Yet, when LR samples are modeled as aggregate conditional means of HR samples with respect to a mediating variable that is globally observed, the recovery of the underlying fine-grained field can be framed as taking an "inverse" of the conditional expectation, namely a deconditioning problem.
2 code implementations • 26 Mar 2021 • David Rindt, Robert Hu, David Steinsaltz, Dino Sejdinovic
We consider frequently used scoring rules for right-censored survival regression models such as time-dependent concordance, survival-CRPS, integrated Brier score and integrated binomial log-likelihood, and prove that neither of them is a proper scoring rule.
no code implementations • 1 Nov 2020 • Tim G. J. Rudner, Dino Sejdinovic, Yarin Gal
We propose Inter-domain Deep Gaussian Processes, an extension of inter-domain shallow GPs that combines the advantages of inter-domain and deep Gaussian processes (DGPs), and demonstrate how to leverage existing approximate inference methods to perform simple and scalable approximate inference using inter-domain features in DGPs.
no code implementations • 23 Aug 2020 • Xingyue Pu, Siu Lun Chau, Xiaowen Dong, Dino Sejdinovic
In this paper, we propose a novel graph learning framework that incorporates the node-side and observation-side information, and in particular the covariates that help to explain the dependency structures in graph signals.
no code implementations • 6 Aug 2020 • Zhu Li, Weijie Su, Dino Sejdinovic
Modern machine learning often operates in the regime where the number of parameters is much higher than the number of data points, with zero training loss and yet good generalization, thereby contradicting the classical bias-variance trade-off.
1 code implementation • 10 Jul 2020 • Anthony Caterini, Rob Cornish, Dino Sejdinovic, Arnaud Doucet
Continuously-indexed flows (CIFs) have recently achieved improvements over baseline normalizing flows on a variety of density estimation tasks.
no code implementations • 6 Jul 2020 • Jean-Francois Ton, Dino Sejdinovic, Kenji Fukumizu
Based on recent developments in meta learning as well as in causal inference, we introduce a novel generative model that allows distinguishing cause and effect in the small data setting.
no code implementations • 2 Jul 2020 • Gustau Camps-Valls, Dino Sejdinovic, Jakob Runge, Markus Reichstein
Earth observation (EO) by airborne and satellite remote sensing and in-situ observations play a fundamental role in monitoring our planet.
no code implementations • 6 Jun 2020 • Siu Lun Chau, Javier González, Dino Sejdinovic
We revisit widely used preferential Gaussian processes by Chu et al.(2005) and challenge their modelling assumption that imposes rankability of data items via latent utility function values.
no code implementations • 8 May 2020 • Siu Lun Chau, Mihai Cucuringu, Dino Sejdinovic
We consider spectral approaches to the problem of ranking n players given their incomplete and noisy pairwise comparisons, but revisit this classical problem in light of player covariate information.
no code implementations • 11 Feb 2020 • Robert Hu, Geoff K. Nicholls, Dino Sejdinovic
We outline an inherent weakness of tensor factorization models when latent factors are expressed as a function of side information and propose a novel method to mitigate this weakness.
1 code implementation • 8 Dec 2019 • Tamara Fernandez, Arthur Gretton, David Rindt, Dino Sejdinovic
We introduce a general non-parametric independence test between right-censored survival times and covariates, which may be multivariate.
no code implementations • 29 Nov 2019 • Duncan Watson-Parris, Samuel Sutherland, Matthew Christensen, Anthony Caterini, Dino Sejdinovic, Philip Stier
One of the most pressing questions in climate science is that of the effect of anthropogenic aerosol on the Earth's energy balance.
no code implementations • 11 Nov 2019 • Zhu Li, Adrian Perez-Suay, Gustau Camps-Valls, Dino Sejdinovic
We present a regularization approach to this problem that trades off predictive accuracy of the learned models (with respect to biased labels) for the fairness in terms of statistical parity, i. e. independence of the decisions from the sensitive covariates.
1 code implementation • 10 Jun 2019 • David Rindt, Dino Sejdinovic, David Steinsaltz
We propose a nonparametric test of independence, termed optHSIC, between a covariate and a right-censored lifetime.
no code implementations • 5 Jun 2019 • Jean-Francois Ton, Lucian Chan, Yee Whye Teh, Dino Sejdinovic
Current meta-learning approaches focus on learning functional representations of relationships between variables, i. e. on estimating conditional expectations in regression.
no code implementations • ICLR 2019 • Jovana Mitrovic, Peter Wirnsberger, Charles Blundell, Dino Sejdinovic, Yee Whye Teh
Infinite-width neural networks have been extensively used to study the theoretical properties underlying the extraordinary empirical success of standard, finite-width neural networks.
no code implementations • 26 Nov 2018 • Francois-Xavier Briol, Chris. J. Oates, Mark Girolami, Michael A. Osborne, Dino Sejdinovic
This article is the rejoinder for the paper "Probabilistic Integration: A Role in Statistical Computation?"
1 code implementation • NeurIPS 2019 • Ho Chung Leon Law, Peilin Zhao, Lucian Chan, Junzhou Huang, Dino Sejdinovic
Bayesian optimisation is a popular technique for hyperparameter learning but typically requires initial exploration even in cases where similar prior tasks have been solved.
2 code implementations • 1 Aug 2018 • Anant Raj, Ho Chung Leon Law, Dino Sejdinovic, Mijung Park
As a result, a simple chi-squared test is obtained, where a test statistic depends on a mean and covariance of empirical differences between the samples, which we perturb for a privacy guarantee.
no code implementations • 6 Jul 2018 • Motonobu Kanagawa, Philipp Hennig, Dino Sejdinovic, Bharath K. Sriperumbudur
This paper is an attempt to bridge the conceptual gaps between researchers working on the two widely used approaches based on positive definite kernels: Bayesian learning or inference using Gaussian processes on the one side, and frequentist kernel methods based on reproducing kernel Hilbert spaces on the other.
no code implementations • 24 Jun 2018 • Zhu Li, Jean-Francois Ton, Dino Oglic, Dino Sejdinovic
We study both the standard random Fourier features method for which we improve the existing bounds on the number of features required to guarantee the corresponding minimax risk convergence rate of kernel ridge regression, as well as a data-dependent modification which samples features proportional to \emph{ridge leverage scores} and further reduces the required number of features.
3 code implementations • NeurIPS 2018 • Anthony L. Caterini, Arnaud Doucet, Dino Sejdinovic
However, for this methodology to be practically efficient, it is necessary to obtain low-variance unbiased estimators of the ELBO and its gradients with respect to the parameters of interest.
1 code implementation • NeurIPS 2018 • Ho Chung Leon Law, Dino Sejdinovic, Ewan Cameron, Tim CD Lucas, Seth Flaxman, Katherine Battle, Kenji Fukumizu
While a typical supervised learning framework assumes that the inputs and the outputs are measured at the same levels of granularity, many applications, including global mapping of disease, only have access to outputs at a much coarser level than that of the inputs.
no code implementations • NeurIPS 2018 • Jovana Mitrovic, Dino Sejdinovic, Yee Whye Teh
Discovering the causal structure among a set of variables is a fundamental problem in many areas of science.
no code implementations • 15 Nov 2017 • Jean-Francois Ton, Seth Flaxman, Dino Sejdinovic, Samir Bhatt
The use of covariance kernels is ubiquitous in the field of spatial statistics.
1 code implementation • 11 May 2017 • Ho Chung Leon Law, Danica J. Sutherland, Dino Sejdinovic, Seth Flaxman
Distribution regression has recently attracted much interest as a generic solution to the problem of supervised learning where labels are available at the group level, rather than at the individual level.
no code implementations • NeurIPS 2017 • Ho Chung Leon Law, Christopher Yau, Dino Sejdinovic
Kernel embeddings of distributions and the Maximum Mean Discrepancy (MMD), the resulting distance between distributions, are useful tools for fully nonparametric two-sample testing and learning on distributions.
3 code implementations • 22 Feb 2017 • Jakob Runge, Dino Sejdinovic, Seth Flaxman
Detecting causal associations in time series datasets is a key challenge for novel insights into complex dynamical systems such as the Earth system or the human brain.
Methodology Atmospheric and Oceanic Physics Applications
no code implementations • 27 Oct 2016 • Seth Flaxman, Yee Whye Teh, Dino Sejdinovic
However, we prove that the representer theorem does hold in an appropriately transformed RKHS, guaranteeing that the optimization of the penalized likelihood can be cast as a tractable finite-dimensional problem.
1 code implementation • 25 Jun 2016 • Qinyi Zhang, Sarah Filippi, Arthur Gretton, Dino Sejdinovic
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.
no code implementations • 30 May 2016 • Gianni Franchi, Jesus Angulo, Dino Sejdinovic
We propose a novel approach for pixel classification in hyperspectral images, leveraging on both the spatial and spectral information in the data.
no code implementations • 7 Mar 2016 • Seth Flaxman, Dino Sejdinovic, John P. Cunningham, Sarah Filippi
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.
no code implementations • 15 Feb 2016 • Jovana Mitrovic, Dino Sejdinovic, Yee Whye Teh
Approximate Bayesian computation (ABC) is an inference framework that constructs an approximation to the true likelihood based on the similarity between the observed and simulated data as measured by a predefined set of summary statistics.
no code implementations • 3 Dec 2015 • François-Xavier Briol, Chris. J. Oates, Mark Girolami, Michael A. Osborne, Dino Sejdinovic
A research frontier has emerged in scientific computation, wherein numerical error is regarded as a source of epistemic uncertainty that can be modelled.
1 code implementation • 11 Oct 2015 • Ingmar Schuster, Heiko Strathmann, Brooks Paige, Dino Sejdinovic
As KSMC does not require access to target gradients, it is particularly applicable on targets whose gradients are unknown or prohibitively expensive.
1 code implementation • NeurIPS 2015 • Kacper Chwialkowski, Aaditya Ramdas, Dino Sejdinovic, Arthur Gretton
The new tests are consistent against a larger class of alternatives than the previous linear-time tests based on the (non-smoothed) empirical characteristic functions, while being much faster than the current state-of-the-art quadratic-time kernel-based or energy distance-based tests.
2 code implementations • NeurIPS 2015 • Heiko Strathmann, Dino Sejdinovic, Samuel Livingstone, Zoltan Szabo, Arthur Gretton
We propose Kernel Hamiltonian Monte Carlo (KMC), a gradient-free adaptive MCMC algorithm based on Hamiltonian Monte Carlo (HMC).
1 code implementation • 9 Mar 2015 • Wittawat Jitkrittum, Arthur Gretton, Nicolas Heess, S. M. Ali Eslami, Balaji Lakshminarayanan, Dino Sejdinovic, Zoltán Szabó
We propose an efficient nonparametric strategy for learning a message operator in expectation propagation (EP), which takes as input the set of incoming messages to a factor node, and produces an outgoing message as output.
no code implementations • 9 Feb 2015 • Mijung Park, Wittawat Jitkrittum, Dino Sejdinovic
Complicated generative models often result in a situation where computing the likelihood of observed data is intractable, while simulating from the conditional density given a parameter value is relatively easy.
no code implementations • 14 Jan 2015 • Heiko Strathmann, Dino Sejdinovic, Mark Girolami
A key quantity of interest in Bayesian inference are expectations of functions with respect to a posterior distribution.
1 code implementation • NeurIPS 2014 • Kacper Chwialkowski, Dino Sejdinovic, Arthur Gretton
A wild bootstrap method for nonparametric hypothesis tests based on kernel distribution embeddings is proposed.
1 code implementation • 19 Jul 2013 • Dino Sejdinovic, Heiko Strathmann, Maria Lomeli Garcia, Christophe Andrieu, Arthur Gretton
A Kernel Adaptive Metropolis-Hastings algorithm is introduced, for the purpose of sampling from a target distribution with strongly nonlinear support.
no code implementations • NeurIPS 2013 • Dino Sejdinovic, Arthur Gretton, Wicher Bergsma
We introduce kernel nonparametric tests for Lancaster three-variable interaction and for total independence, using embeddings of signed measures into a reproducing kernel Hilbert space.
no code implementations • NeurIPS 2012 • Arthur Gretton, Dino Sejdinovic, Heiko Strathmann, Sivaraman Balakrishnan, Massimiliano Pontil, Kenji Fukumizu, Bharath K. Sriperumbudur
A means of parameter selection for the two-sample test based on the MMD is proposed.
no code implementations • 25 Jul 2012 • Dino Sejdinovic, Bharath Sriperumbudur, Arthur Gretton, Kenji Fukumizu
We provide a unifying framework linking two classes of statistics used in two-sample and independence testing: on the one hand, the energy distances and distance covariances from the statistics literature; on the other, maximum mean discrepancies (MMD), that is, distances between embeddings of distributions to reproducing kernel Hilbert spaces (RKHS), as established in machine learning.