Moritz Schauer
YOU?
Author Swipe
View article: Backward Filtering Forward Guiding
Backward Filtering Forward Guiding Open
We develop a general methodological framework for probabilistic inference in discrete- and continuous-time stochastic processes evolving on directed acyclic graphs (DAGs). The process is observed only at the leaf nodes, and the challenge i…
View article: Rebalancing Markov jump processes for non-reversible continuous-time sampling
Rebalancing Markov jump processes for non-reversible continuous-time sampling Open
Markov chain Monte Carlo methods are central in computational statistics, and typically rely on detailed balance to ensure invariance with respect to a target distribution. Although straightforward to construct by Metropolization, this can…
View article: Controlled stochastic processes for simulated annealing
Controlled stochastic processes for simulated annealing Open
Simulated annealing solves global optimization problems by means of a random walk in a cooling energy landscape based on the objective function and a temperature parameter. However, if the temperature is decreased too quickly, this procedu…
View article: Guided smoothing and control for diffusion processes
Guided smoothing and control for diffusion processes Open
The smoothing distribution is the conditional distribution of the diffusion process in the space of trajectories given noisy observations made continuously in time. It is generally difficult to sample from this distribution. We use the the…
View article: Score matching for bridges without learning time-reversals
Score matching for bridges without learning time-reversals Open
We propose a new algorithm for learning bridged diffusion processes using score-matching methods. Our method relies on reversing the dynamics of the forward process and using this to learn a score function, which, via Doob's $h$-transform,…
View article: Gradient Estimation via Differentiable Metropolis-Hastings
Gradient Estimation via Differentiable Metropolis-Hastings Open
Metropolis-Hastings estimates intractable expectations - can differentiating the algorithm estimate their gradients? The challenge is that Metropolis-Hastings trajectories are not conventionally differentiable due to the discrete accept/re…
View article: Simulating conditioned diffusions on manifolds
Simulating conditioned diffusions on manifolds Open
To date, most methods for simulating conditioned diffusions are limited to the Euclidean setting. The conditioned process can be constructed using a change of measure known as Doob's $h$-transform. The specific type of conditioning depends…
View article: Causal structure learning with momentum: Sampling distributions over Markov Equivalence Classes of DAGs
Causal structure learning with momentum: Sampling distributions over Markov Equivalence Classes of DAGs Open
In the context of inferring a Bayesian network structure (directed acyclic graph, DAG for short), we devise a non-reversible continuous time Markov chain, the ``Causal Zig-Zag sampler'', that targets a probability distribution over classes…
View article: Differentiating Metropolis-Hastings to Optimize Intractable Densities
Differentiating Metropolis-Hastings to Optimize Intractable Densities Open
We develop an algorithm for automatic differentiation of Metropolis-Hastings samplers, allowing us to differentiate through probabilistic inference, even if the model has discrete components within it. Our approach fuses recent advances in…
View article: Compositionality in algorithms for smoothing
Compositionality in algorithms for smoothing Open
Backward Filtering Forward Guiding (BFFG) is a bidirectional algorithm proposed in Mider et al. [2021] and studied more in depth in a general setting in Van der Meulen and Schauer [2022]. In category theory, optics have been proposed for m…
View article: Methods and applications of PDMP samplers with boundary conditions
Methods and applications of PDMP samplers with boundary conditions Open
We extend Monte Carlo samplers based on piecewise deterministic Markov processes (PDMP samplers) by formally defining different boundary conditions such as sticky floors, soft and hard walls and teleportation portals. This allows PDMP samp…
View article: Flexible Group Fairness Metrics for Survival Analysis
Flexible Group Fairness Metrics for Survival Analysis Open
Purpose : Algorithmic fairness is an increasingly important field concerned with detecting and mitigating biases in machine learning models. There has been a wealth of literature for algorithmic fairness in regression and classification ho…
View article: Conditioning continuous-time Markov processes by guiding
Conditioning continuous-time Markov processes by guiding Open
A continuous-time Markov process X can be conditioned to be in a given state at a fixed time T>0 using Doob's h-transform. This transform requires the typically intractable transition density of X. The effect of the h-transform can be d…
View article: Automatic Differentiation of Programs with Discrete Randomness
Automatic Differentiation of Programs with Discrete Randomness Open
Automatic differentiation (AD), a technique for constructing new programs which compute the derivative of an original program, has become ubiquitous throughout scientific computing and deep learning due to the improved performance afforded…
View article: Nonparametric Bayesian volatility estimation for gamma-driven stochastic differential equations
Nonparametric Bayesian volatility estimation for gamma-driven stochastic differential equations Open
We study a nonparametric Bayesian approach to estimation of the volatility function of a stochastic differential equation driven by a gamma process. The volatility function is modelled a priori as piecewise constant, and we specify a gamma…
View article: Sticky PDMP samplers for sparse and local inference problems
Sticky PDMP samplers for sparse and local inference problems Open
We construct a new class of efficient Monte Carlo methods based on continuous-time piecewise deterministic Markov processes (PDMPs) suitable for inference in high dimensional sparse models, i.e. models for which there is prior knowledge th…
View article: Applied Measure Theory for Probabilistic Modeling
Applied Measure Theory for Probabilistic Modeling Open
Probabilistic programming and statistical computing are vibrant areas in the development of the Julia programming language, but the underlying infrastructure dramatically predates recent developments. The goal of MeasureTheory.jl is to pro…
View article: Applied Measure Theory for Probabilistic Modeling
Applied Measure Theory for Probabilistic Modeling Open
Probabilistic programming and statistical computing are vibrant areas in the development of the Julia programming language, but the underlying infrastructure dramatically predates recent developments. The goal of MeasureTheory.jl is to pro…
View article: Flexible Group Fairness Metrics for Survival Analysis
Flexible Group Fairness Metrics for Survival Analysis Open
Algorithmic fairness is an increasingly important field concerned with detecting and mitigating biases in machine learning models. There has been a wealth of literature for algorithmic fairness in regression and classification however ther…
View article: Diffusion Bridges for Stochastic Hamiltonian Systems and Shape Evolutions
Diffusion Bridges for Stochastic Hamiltonian Systems and Shape Evolutions Open
Stochastically evolving geometric systems are studied in shape analysis and computational anatomy for modeling random evolutions of human organ shapes. The notion of geodesic paths between shapes is central to shape analysis and has a natu…
View article: Conditioning continuous-time Markov processes by guiding
Conditioning continuous-time Markov processes by guiding Open
A continuous-time Markov process $X$ can be conditioned to be in a given state at a fixed time $T > 0$ using Doob's $h$-transform. This transform requires the typically intractable transition density of $X$. The effect of the $h$-transform…
View article: Weak solutions to gamma-driven stochastic differential equations
Weak solutions to gamma-driven stochastic differential equations Open
We study a stochastic differential equation driven by a gamma process, for which we give results on the existence of weak solutions under conditions on the volatility function. To that end we provide results on the density process between …
View article: JuliaPlots/Makie.jl: v0.15.0
JuliaPlots/Makie.jl: v0.15.0 Open
Makie v0.15.0 Diff since v0.14.2 Closed issues: Segmentation fault for nested plots (#540) update_cam! doesn't respect the value of Camera3D.far (#801) Camera weirdness (#941) CHANGELOG.md confusing (#961) linewidth argument per line in li…
View article: mschauer/Bridge.jl: v0.11.6
mschauer/Bridge.jl: v0.11.6 Open
Bridge v0.11.6 Diff since v0.11.5 Closed issues: Release 0.11.* (#65) Merged pull requests: CompatHelper: bump compat for "StaticArrays" to "1.0" (#84) (@github-actions[bot]) CompatHelper: bump compat for "SpecialFunctions" to "1.2" (#86) …
View article: Sticky PDMP samplers for sparse and local inference problems
Sticky PDMP samplers for sparse and local inference problems Open
We construct a new class of efficient Monte Carlo methods based on continuous-time piecewise deterministic Markov processes (PDMPs) suitable for inference in high dimensional sparse models, i.e. models for which there is prior knowledge th…
View article: chrisvoncsefalvay/learn-julia-the-hard-way: v0.85: February 2020
chrisvoncsefalvay/learn-julia-the-hard-way: v0.85: February 2020 Open
Learn Julia the hard way!