arXiv research

Differentiable relaxation for inferring partial orders from noisy linear data.

Algorithm improves reinforcement learning in MDPs with partial order policies.

Hierarchical Partial-Order Models for Ranking

Paper proposes methods to learn DAGs from partial orderings.

A new method for state estimation on complex networks.

Neural autoregressive models are explicit density estimators that achieve state-of-the-art likelihoods for generative modeling. The D-dimensional data distribution is factorized into an autoregressive product of one-dimensional conditional distributions according to the chain rule. Data completion is a more involved ta…

Sig-DEG speeds up diffusion models by distilling them into faster approximations.

WNVI solves inverse problems without forward models using neural networks.

The paper tackles partial inference in structured prediction using a convex optimization approach.

Ribbon cobordism forms a partial order in 3-manifolds.

Inference for normal and Monte Carlo distributions using minimum relative entropy.

Directed graphs occur throughout statistical modeling of networks, and exchangeability is a natural assumption when the ordering of vertices does not matter. There is a deep structural theory for exchangeable undirected graphs, which extends to the directed case via measurable objects known as digraphons. Using digraph…

We describe Venture, an interactive virtual machine for probabilistic programming that aims to be sufficiently expressive, extensible, and efficient for general-purpose use. Like Church, probabilistic models and inference problems in Venture are specified via a Turing-complete, higher-order probabilistic language desce…

Extends causal discovery to group variables, improving performance in real-world applications.

Ribbon cobordisms form a partial order on 3-manifolds.

New method for inferring network topology from partial data.

In reinforcement learning (RL), agents often operate in partially observed and uncertain environments. Model-based RL suggests that this is best achieved by learning and exploiting a probabilistic model of the world. 'Active inference' is an emerging normative framework in cognitive and computational neuroscience that …

New BED method handles online inference for partially observed dynamical systems.

New algorithm infers trajectories from partial observations using optimal transport.

Prove strong ribbon concordance induces a partial order on links, certify minimality for a handful of knots, and find minimal ribbon minimal knots.

Rational homology ribbon cobordism defines a partial order on 3-manifolds.

Study uses deep neural networks for inference in partially linear models with dependent data.

A new approach uses partial likelihood to improve tree-based density estimation and inference.

The paper uses belief propagation to analyze rankings and partial orders from partial information.

ParKCa combines multiple causal inference methods to infer new causes from known and unknown factors.

New methods improve estimation accuracy in noisy settings.

PPI uses predictions and weighting to infer from partially labeled data.

This work leverages recent advances in probabilistic machine learning to discover conservation laws expressed by parametric linear equations. Such equations involve, but are not limited to, ordinary and partial differential, integro-differential, and fractional order operators. Here, Gaussian process priors are modifie…

In this work, we propose a simple yet effective solution to the problem of connectome inference in calcium imaging data. The proposed algorithm consists of two steps. First, processing the raw signals to detect neural peak activities. Second, inferring the degree of association between neurons from partial correlation …

This paper accelerates inverse solutions for PDEs using ML and ROMs.

Efficiently learns from partial labels using variational inference.

In this paper, we introduce a partial order on neighborhood equivalence classes of maximally spread essential multibranched surfaces embedded in a 3-manifold. We show that if a maximally spread essential multibranched surface is atoroidal and acylindrical, then its equivalence class is minimal with respect to the parti…

It is well known that a countable group admits a left-invariant total order if and only if it acts faithfully on R by orientation preserving homeomorphisms. Such group actions are special cases of group actions on simply connected 1-manifolds, or equivalently, actions on oriented order trees. We characterize a class of…

Transformer learns to infer partial MDPs for efficient in-context adaptation and exploration.

Develops statistical methods for rates of change on Riemannian manifolds.

Develops a method to infer partial rankings from sparse comparisons.

Paper proposes efficient weight updates for edge nodes with minimal communication.

Efficiently predicts high-fidelity PDE solutions using multi-fidelity Gaussian processes.

New method trains reflected Schrödinger bridges without complex derivatives.

New statistical models for predicting ranked preferences from partial orders.

We present a new Markov chain Monte Carlo method for estimating posterior probabilities of structural features in Bayesian networks. The method draws samples from the posterior distribution of partial orders on the nodes; for each sampled partial order, the conditional probabilities of interest are computed exactly. We…

Improved Bayesian neural network inference by selectively removing redundant modes.

A new RG approach connects discrete and continuous time descriptions of Gaussian processes.

A novel transformer model improves classification of partially ordered sequences.

Unified method for inference on partially identified causal effects using covariates.

Temporal Point Processes (TPP) with partial likelihoods involving a latent structure often entail an intractable marginalization, thus making inference hard. We propose a novel approach to Maximum Likelihood Estimation (MLE) involving approximate inference over the latent variables by minimizing a tight upper bound on …

Belief Propagation has been widely used for marginal inference, however it is slow on problems with large-domain variables and high-order factors. Previous work provides useful approximations to facilitate inference on such models, but lacks important anytime properties such as: 1) providing accurate and consistent mar…

Deep active inference learns policies from sensory inputs.