Michael U. Gutmann
YOU?
Author Swipe
CFMI: Flow Matching for Missing Data Imputation Open
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 intractabilit…
View article: Risk-averse optimization of genetic circuits under uncertainty
Risk-averse optimization of genetic circuits under uncertainty Open
Synthetic biology aims to engineer biological systems with specified functions. This requires navigating an extensive design space, which is challenging to achieve with wet lab experiments alone. To expedite the design process, mathematica…
Improving Variational Autoencoder Estimation from Incomplete Data with Mixture Variational Families Open
We consider the task of estimating variational autoencoders (VAEs) when the training data is incomplete. We show that missing data increases the complexity of the model's posterior distribution over the latent variables compared to the ful…
Designing optimal behavioral experiments using machine learning Open
Computational models are powerful tools for understanding human cognition and behavior. They let us express our theories clearly and precisely and offer predictions that can be subtle and often counter-intuitive. However, this same richnes…
An Extendable <i>Python</i> Implementation of Robust Optimization Monte Carlo Open
Performing inference in statistical models with an intractable likelihood is challenging, therefore, most likelihood-free inference (LFI) methods encounter accuracy and efficiency limitations. In this paper, we present the implementation o…
An Extendable Python Implementation of Robust Optimisation Monte Carlo Open
Performing inference in statistical models with an intractable likelihood is challenging, therefore, most likelihood-free inference (LFI) methods encounter accuracy and efficiency limitations. In this paper, we present the implementation o…
Conditional Sampling of Variational Autoencoders via Iterated Approximate Ancestral Sampling Open
Conditional sampling of variational autoencoders (VAEs) is needed in various applications, such as missing data imputation, but is computationally intractable. A principled choice for asymptotically exact conditional sampling is Metropolis…
Designing Optimal Behavioral Experiments Using Machine Learning Open
Computational models are powerful tools for understanding human cognition and behavior. They let us express our theories clearly and precisely, and offer predictions that can be subtle and often counter-intuitive. However, this same richne…
Estimating the Density Ratio between Distributions with High Discrepancy using Multinomial Logistic Regression Open
Functions of the ratio of the densities $p/q$ are widely used in machine learning to quantify the discrepancy between the two distributions $p$ and $q$. For high-dimensional distributions, binary classification-based density ratio estimato…
View article: Systematic comparison of ranking aggregation methods for gene lists in experimental results
Systematic comparison of ranking aggregation methods for gene lists in experimental results Open
Motivation A common experimental output in biomedical science is a list of genes implicated in a given biological process or disease. The gene lists resulting from a group of studies answering the same, or similar, questions can be combine…
Bayesian Optimization with Informative Covariance Open
Bayesian optimization is a methodology for global optimization of unknown and expensive objectives. It combines a surrogate Bayesian regression model with an acquisition function to decide where to evaluate the objective. Typical regressio…
Enhanced gradient-based MCMC in discrete spaces Open
The recent introduction of gradient-based MCMC for discrete spaces holds great promise, and comes with the tantalising possibility of new discrete counterparts to celebrated continuous methods such as MALA and HMC. Towards this goal, we in…
Inference and uncertainty quantification of stochastic gene expression via synthetic models Open
Estimating uncertainty in model predictions is a central task in quantitative biology. Biological models at the single-cell level are intrinsically stochastic and nonlinear, creating formidable challenges for their statistical estimation w…
Pen and Paper Exercises in Machine Learning Open
This is a collection of (mostly) pen-and-paper exercises in machine learning. The exercises are on the following topics: linear algebra, optimisation, directed graphical models, undirected graphical models, expressive power of graphical mo…
Statistical applications of contrastive learning Open
The likelihood function plays a crucial role in statistical inference and experimental design. However, it is computationally intractable for several important classes of statistical models, including energy-based models and simulator-base…
Inference and Uncertainty Quantification of Stochastic Gene Expression via Synthetic Models Open
A bstract Estimating uncertainty in model predictions is a central task in quantitative biology. Biological models at the single-cell level are intrinsically stochastic and nonlinear, creating formidable challenges for their statistical es…
View article: Systematic comparison of ranking aggregation methods for gene lists in experimental results
Systematic comparison of ranking aggregation methods for gene lists in experimental results Open
A common experimental output in biomedical science is a list of genes implicated in a given biological process or disease. The results of a group of studies answering the same, or similar, questions can be combined by meta-analysis to find…
Variational Gibbs Inference for Statistical Model Estimation from\n Incomplete Data Open
Statistical models are central to machine learning with broad applicability\nacross a range of downstream tasks. The models are controlled by free\nparameters that are typically estimated from data by maximum-likelihood\nestimation or appr…
Variational Gibbs Inference for Statistical Model Estimation from Incomplete Data Open
Statistical models are central to machine learning with broad applicability across a range of downstream tasks. The models are controlled by free parameters that are typically estimated from data by maximum-likelihood estimation or approxi…
Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods Open
We introduce implicit Deep Adaptive Design (iDAD), a new method for performing adaptive experiments in real-time with implicit models. iDAD amortizes the cost of Bayesian optimal experimental design (BOED) by learning a design policy netwo…
Bayesian Optimal Experimental Design for Simulator Models of Cognition Open
Bayesian optimal experimental design (BOED) is a methodology to identify experiments that are expected to yield informative data. Recent work in cognitive science considered BOED for computational models of human behavior with tractable an…
Gradient-based Bayesian Experimental Design for Implicit Models using Mutual Information Lower Bounds Open
We introduce a framework for Bayesian experimental design (BED) with implicit models, where the data-generating distribution is intractable but sampling from it is still possible. In order to find optimal experimental designs for such mode…
Bayesian Experimental Design for Intractable Models of Cognition Open
Bayesian experimental design (BED) is a methodology to identify designs that are expected to yield informative data. Recent work in cognitive science considered BED for cognitive models with tractable and known likelihood functions. Howeve…
Extending the statistical software package Engine for Likelihood-Free\n Inference Open
Bayesian inference is a principled framework for dealing with uncertainty.\nThe practitioner can perform an initial assumption for the physical phenomenon\nthey want to model (prior belief), collect some data and then adjust the\ninitial a…
Extending the statistical software package Engine for Likelihood-Free Inference Open
Bayesian inference is a principled framework for dealing with uncertainty. The practitioner can perform an initial assumption for the physical phenomenon they want to model (prior belief), collect some data and then adjust the initial assu…
Stir to Pour: Efficient Calibration of Liquid Properties for Pouring Actions Open
Humans use simple probing actions to develop intuition about the physical behaviour of common objects. Such intuition is particularly useful for adaptive estimation of favourable manipulation strategies of those objects in novel contexts. …
Neural Approximate Sufficient Statistics for Implicit Models Open
We consider the fundamental problem of how to automatically construct summary statistics for implicit generative models where the evaluation of the likelihood function is intractable, but sampling data from the model is possible. The idea …
Likelihood-Free Inference by Ratio Estimation Open
We consider the problem of parametric statistical inference when likelihood computations are prohibitively expensive but sampling from the model is possible. Several so-called likelihood-free methods have been developed to perform inferenc…
Sequential Bayesian Experimental Design for Implicit Models via Mutual Information Open
Bayesian experimental design (BED) is a framework that uses statistical models and decision making under uncertainty to optimise the cost and performance of a scientific experiment. Sequential BED, as opposed to static BED, considers the s…