arXiv research

Reduces IB problem to a simpler, lower-dimensional problem.

New method uses diffusion models for Bayesian inverse problems.

BCD Nets use variational inference to estimate DAGs with uncertainty.

Paper proposes a new Taylor moment expansion for non-linear Gaussian filtering and smoothing.

Gaussian processes are conditioned on various types of data.

Local approach learns causal structure of linear Gaussian polytree models from interventional data.

Paper studies sparsity and DAG constraints for learning linear DAGs.

The NL score optimizes speaker recognition tasks.

Method predicts future rewards from past actions in a linear Gaussian system.

Framework for sensitivity analysis in biomanufacturing processes.

Jump Markov linear models consists of a finite number of linear state space models and a discrete variable encoding the jumps (or switches) between the different linear models. Identifying jump Markov linear models makes for a challenging problem lacking an analytical solution. We derive a new expectation maximization …

Using the linear Gaussian latent variable model as a starting point we relax some of the constraints it imposes by deriving a nonparametric latent feature Gaussian variable model. This model introduces additional discrete latent variables to the original structure. The Bayesian nonparametric nature of this new model al…

Bayesian networks are typically faithful, with implications for causal inference.

Generalizes bits back coding for time-series models with latent Markov structures.

Proposes NUV priors for half-space and box constraints.

New method forecasts time series with changing variances.

Study the properties of SGD in non-vanishing learning rate regime.

Test-time training adapts a pretrained model to each prompt via parameter updates, improving accuracy under pretraining-to-test distribution shifts.

Study non-Gaussian measures' concentration properties in metric spaces.

Proposes GPLFR for predicting high-dimensional outputs with few data.

Over the past two decades, several consistent procedures have been designed to infer causal conclusions from observational data. We prove that if the true causal network might be an arbitrary, linear Gaussian network or a discrete Bayes network, then every unambiguous causal conclusion produced by a consistent method f…

We report an exact likelihood computation for Linear Gaussian Markov processes that is more scalable than existing algorithms for complex models and sparsely sampled signals. Better scaling is achieved through elimination of repeated computations in the Kalman likelihood, and by using the diagonalized form of the state…

New algorithm estimates causal effects for non-Gaussian data.

A model learns causal representations from high-dimensional data.

In this paper we introduce Feature Gradients, a gradient-based search algorithm for feature selection. Our approach extends a recent result on the estimation of learnability in the sublinear data regime by showing that the calculation can be performed iteratively (i.e., in mini-batches) and in linear time and space wit…

This study uses HMM and RL to dynamically allocate equities, Treasuries, and gold based on market regimes.

We propose a novel time discretization for the log-normal SABR model and derive its asymptotic properties.

This work models market regimes using CTMSTOU and simulates trading policies.

Building on the work of Schweizer (1995) and Cern and Kallseny (2007), we present discrete time formulas minimizing the mean square hedging error for multidimensional assets. In particular, we give explicit formulas when a regime-switching random walk or a GARCH-type process is utilized to model the returns. Monte Carl…

Unified framework for inference in complex nonlinear processes.

Bayesian active learning reduces data needed for latent variable models.

The paper extends invariant theory to non-compact and non-reductive actions, classifying four regimes.

A new framework for efficient sequence maps using Bayesian filtering and covariance.

We study the information-theoretic lower bound of the sample complexity of the correct recovery of diffusion network structures. We introduce a discrete-time diffusion model based on the Independent Cascade model for which we obtain a lower bound of order $Ω(k \log p)$, for directed graphs of $p$ nodes, and at most $k$…

New algorithms improve sampling from complex distributions.

New model identifies regimes in non-stationary data.

In this paper, we consider the problem of pricing discretely-sampled variance swaps based on a hybrid model of stochastic volatility and stochastic interest rate with regime-switching. Our modelling framework extends the Heston stochastic volatility model by including the CIR stochastic interest rate and model paramete…

Gaussian belief propagation (BP) has been widely used for distributed inference in large-scale networks such as the smart grid, sensor networks, and social networks, where local measurements/observations are scattered over a wide geographical area. One particular case is when two neighboring agents share a common obser…

In this paper, we consider a discrete time economy where we assume that the short term interest rate follows a quadratic term structure of a regime switching asset process. The possible non-linear structure and the fact that the interest rate can have different economic or financial trends justify the interest of Regim…

This work extends identifiability analysis to sequential latent variable models, focusing on Switching Dynamical Systems.

Critically ill patients in regular wards are vulnerable to unanticipated clinical dete- rioration which requires timely transfer to the intensive care unit (ICU). To allow for risk scoring and patient monitoring in such a setting, we develop a novel Semi- Markov Switching Linear Gaussian Model (SSLGM) for the inpatient…

The study sets limits on how well systems can be controlled adaptively.

The study identifies and analyzes different market regimes in equity markets using advanced signal processing techniques.

New method clusters financial time series into volatility regimes.

Many applications, including natural language processing, sensor networks, collaborative filtering, and federated learning, call for estimating discrete distributions from data collected in batches, some of which may be untrustworthy, erroneous, faulty, or even adversarial. Previous estimators for this setting ran in e…

New method for estimating treatment effects without complex propensity models.

Neural networks learn discrete tasks on continuous data via emergent geometry.

We propose a correlated stochastic process of which the novel non-Gaussian probability mass function is constructed by exactly solving moment generating function. The calculation of cumulants and auto-correlation shows that the process is convergent and scale invariant in the large but finite number limit. We demonstra…