We consider off-policy policy evaluation when the trajectory data are generated by multiple behavior policies. Recent work has shown the key role played by the state or state-action stationary distribution corrections in the infinite horizon context for off-policy policy evaluation. We propose estimated mixture policy …
On-device research index
arXiv research
A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
168,695 papers · 148 categories
Trend · papers per month
48 results for “stationary distribution correction”
Unified DICE estimators as regularized Lagrangians for improved off-policy evaluation.
problem Improving off-policy evaluation from behavior-agnostic data.
method Unified derivation of DICE estimators as regularized Lagrangians of a linear program.
result Dual solutions offer greater flexibility and provide superior estimates in practice.
Improved Adam for time series forecasting with distributional drift.
problem Non-stationary data challenges Adam's effectiveness.
method Proposed TS_Adam, removing Adam's second-order bias correction.
result TS_Adam achieves 12.8% reduction in MSE and 5.7% in MAE on ETT datasets.
GenDICE estimates stationary values from offline data.
problem Estimating stationary values from limited offline data.
method Ratio correction based on stationary distribution properties, variational divergence minimization.
result Consistent estimation of stationary values possible in offline settings.
Estimates stationary distribution from batch transitions without access to the underlying process.
problem Estimating stationary distribution from batch transitions without access to the underlying process.
method Proposes a consistent estimator based on a correction ratio function and variational power method (VPM).
result VPM provides significantly better estimates across various problems.
BC-ACI corrects time series forecast bias, improving prediction intervals.
problem Persistent bias in time series forecasts leads to overly conservative prediction intervals.
method Augments ACI with an EWM estimate of forecast bias to correct nonconformity scores and re-center intervals.
result Reduces Winkler interval scores by 13-17% under distribution shifts, improving calibration.
In many real-world reinforcement learning applications, access to the environment is limited to a fixed dataset, instead of direct (online) interaction with the environment. When using this data for either evaluation or training of a new policy, accurate estimates of discounted stationary distribution ratios -- correct…
This research improves online learning by correcting for target shift in machine learning.
problem Online learning struggles with distributional shift, especially in target values.
method Derives closed-form expressions for online and offline learning, and target correction.
result Online kernel-based learning can learn the same predictor as offline learning with target correction.
Ruin Probabilities for Risk Processes with Non-Stationary Arrivals and Subexponential Claimsq-fin.RM
In this paper, we obtain the finite-horizon and infinite-horizon ruin probability asymptotics for risk processes with claims of subexponential tails for non-stationary arrival processes that satisfy a large deviation principle. As a result, the arrival process can be dependent, non-stationary and non-renewal. We give t…
FedSplit improves federated learning by ensuring correct convergence to optimal solutions.
problem Federated learning's fixed points do not always correspond to optimal solutions in simple convex settings.
method FedSplit uses operator splitting procedures to solve distributed convex minimization problems with additive structure.
result FedSplit ensures that the fixed points correspond to optima of the original optimization problem.
Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneitycs.LG
Rescaled ASGD optimizes distributed learning under heterogeneous data.
problem Vanilla ASGD biases towards a frequency-weighted average of local objectives.
method Rescale worker stepsizes by their computation times.
result Rescaled ASGD converges to the correct global objective in fixed-computation model.
Algorithm learns stochastic system dynamics from data.
problem Recovering interpretable symbolic expressions for stochastic systems.
method Data-driven, trajectory averaging, drift-informed correction.
result Recover coefficients and densities to within 5% and 0.01 in total variation, respectively.
New GP kernels avoid mean reversion without losing smoothness.
problem Pathological behavior in stationary GP regression.
method Improper Gaussian processes with non-positive kernels.
result Stationary, non-reverting covariance functions.
The method approximates stationary distributions of Markov models by truncating irrelevant states.
problem Computing the stationary distribution of complex Markov models is computationally challenging.
method A state-space lumping scheme that aggregates states in a grid structure, iteratively refining the state-space.
result The method provides a well-justified finite-state projection tailored to the stationary behavior of Markov models.
Proposes a method to forecast non-stationary time series.
problem Challenges of non-stationary conditional distributions in deep learning.
method Bayesian dynamic model + deep conditional distribution model.
result Adapts to non-stationary time series better than state-of-the-art solutions.
Hamiltonian Monte Carlo is a powerful algorithm for sampling from difficult-to-normalize posterior distributions. However, when the geometry of the posterior is unfavorable, it may take many expensive evaluations of the target distribution and its gradient to converge and mix. We propose neural transport (NeuTra) HMC, …
The Probably Approximately Correct (PAC) Bayes framework (McAllester, 1999) can incorporate knowledge about the learning algorithm and (data) distribution through the use of distribution-dependent priors, yielding tighter generalization bounds on data-dependent posteriors. Using this flexibility, however, is difficult,…
New method improves sample diversity and efficiency from complex distributions.
problem Sampling from intractable un-normalized distributions with high auto-correlation.
method Stein self-repulsive dynamics using a repulsive force to push samples away from past trajectories.
result Significantly decreases auto-correlation and increases effective sample size.
The problem of change-point estimation is considered under a general framework where the data are generated by unknown stationary ergodic process distributions. In this context, the consistent estimation of the number of change-points is provably impossible. However, it is shown that a consistent clustering method may …
Exact Multivariate Amplitude Distributions for Non-Stationary Gaussian or Algebraic Fluctuations of Covariances or Correlationscond-mat.dis-nn
The paper provides exact multivariate amplitude distributions for non-stationary Gaussian or algebraic fluctuations.
problem Capturing the statistical properties of fluctuating correlations in non-stationary systems.
method Developed a random matrix model to average multivariate amplitude distributions from short time scales to large time scales.
result Explicit multivariate distributions for non-stationary correlation systems are provided, capturing the degree of non-stationarity.
Innovative game theory approach optimizes survival analysis metrics.
problem Survival analysis models trained with maximum likelihood do not directly optimize criteria like Brier score or Bernoulli log likelihood.
method Inverse-Weighted Survival Games: Construct objectives from re-weighted estimates featuring the other model, holding the latter fixed during training.
result Games optimize Brier score on simulations and real-world data.
Recent work has argued that stochastic gradient descent can approximate the Bayesian uncertainty in model parameters near local minima. In this work we develop a similar correspondence for minibatch natural gradient descent (NGD). We prove that for sufficiently small learning rates, if the model predictions on the trai…
Improves decision tree performance by correcting split selection errors.
problem Invalid statistical guarantees in split selection for decision trees.
method Introduces anytime-valid inference to provide valid statistical guarantees.
result Provides anytime-valid control of false splits under arbitrary data streams.
We study theoretical properties of regularized robust M-estimators, applicable when data are drawn from a sparse high-dimensional linear model and contaminated by heavy-tailed distributions and/or outliers in the additive errors and covariates. We first establish a form of local statistical consistency for the penalize…
Improved sampling from mean-field stationary distributions.
problem Sampling from the stationary distribution of mean-field SDEs.
method Decoupling the problem into two aspects: approximation of mean-field SDE and sampling from finite-particle distribution.
result Improved guarantees in various settings, including optimizing neural networks.
New method reduces computational cost for learning stationary diffusions.
problem Learning parameters of stationary diffusions efficiently.
method Stein-type discrepancy (SKDS) for estimating generator expectations.
result SKDS guarantees alignment with target stationary distribution.
It is well-known that irreversible MCMC algorithms converge faster to their stationary distributions than reversible ones. Using the special geometric structure of Lie groups and dissipation fields compatible with the symplectic structure, we construct an irreversible HMC-like MCMC algorithm on $\mathcal G…
Study of recurrences in earthquakes, climate, financial time-series, etc. is crucial to better forecast disasters and limit their consequences. However, almost all the previous phenomenological studies involved only a long-ranged autocorrelation function, or disregarded the multi-scaling properties induced by potential…
New SDE model from machine learning optimization with unique stationary distribution.
problem Stationary distribution of machine learning optimization models.
method Proved ergodicity and unique stationary distribution of power-law dynamic SDE.
result Power-law dynamic has a unique stationary distribution and is ergodic.
The Langevin Algorithm's stationary distribution is shown to be sub-exponential or sub-Gaussian under certain conditions.
problem Understanding the properties of the Langevin Algorithm's stationary distribution.
method Analysis using a rotation-invariant moment generating function (Bessel function) to study the stationary dynamics of the Langevin Algorithm.
result Concentration results for the Langevin Algorithm's stationary distribution are established, showing it is sub-exponential or sub-Gaussian under convex or strongly convex potential conditions.
PyChEst detects changes in non-stationary time series without distributional assumptions.
problem Detecting changes in non-stationary time series data.
method Nonparametric algorithms for consistent detection of multiple changepoints in piece-wise stationary processes.
result PyChEst consistently detects changes without distributional assumptions.
Study reveals convergence properties of SGD with random learning rate.
problem Analyzing convergence of SGD with random learning rate in non-convex optimization.
method Introduced Poisson SGD with random learning rate and used stationary distribution analysis.
result Poisson SGD converges to a stationary distribution and finds global minima in non-convex optimization.
We model non-stationary volume-price distributions with a log-normal distribution and collect the time series of its two parameters. The time series of the two parameters are shown to be stationary and Markov-like and consequently can be modelled with Langevin equations, which are derived directly from their series of …
New definition resolves ambiguity in non-stationary bandit classification.
problem Ambiguity in classifying non-stationary bandits using existing definitions.
method Introducing a formal definition that resolves ambiguity and provides a unified approach.
result Unified approach applicable to both Bayesian and frequentist formulations, resolves classification issues.
The paper develops a stationary-distribution theory for Random Forest ensemble size selection.
problem Determining the optimal number of trees in Random Forests.
method Modeling the ensemble size as a birth-death Markov chain and deriving its stationary distribution.
result The stationary ensemble size scales as as .
The paper studies Gauss maps of space-like stationary surfaces in Lorentz-Minkowski space, focusing on ramification and unicity.
problem Value distribution properties of Gauss maps on space-like stationary surfaces.
method Investigation of ramification and unicity properties, considering rational graphic Gauss images.
result Obtained general conclusions similar to Euclidean space, extending to rational graphic Gauss images.
Empirical Transfer Operators and Finite-Sample Change Detection for Noisy Expanding Interval Mapsstat.ML
Detect changes in noisy dynamical systems using empirical approximations and finite-sample bounds.
problem Change detection in noisy dynamical systems
method Partition-based empirical approximations and finite-state stationary distribution stability
result Finite-sample bound for empirical stationary density
A defining feature of non-stationary systems is the time dependence of their statistical parameters. Measured time series may exhibit Gaussian statistics on short time horizons, due to the central limit theorem. The sample statistics for long time horizons, however, averages over the time-dependent parameters. To model…
Non-Stationary Covariance Matrices And Noisephysics.soc-ph
The exact meaning of the noise spectrum of eigenvalues of the covariance matrix is discussed. In order to better understand the possible phenomena behind the observed noise, the spectrum of eigenvalues of the covariance matrix is studied under a model where most of the true eigenvalues are zero and the parameters are n…
The problem of time-series clustering is considered in the case where each data-point is a sample generated by a piecewise stationary ergodic process. Stationary processes are perhaps the most general class of processes considered in non-parametric statistics and allow for arbitrary long-range dependence between variab…
Improved method finds second-order stationary points privately with better efficiency.
problem Finding second-order stationary points privately under differential privacy constraints.
method Adaptive batch sizes and binary tree mechanism.
result Improved bound for privately finding SOSP, matching state-of-the-art for FOSP.
Study on fake stationary Volterra Heston model for non-stationary processes.
problem Non-stationary nature of true Volterra equations.
method Weak notion of stationarity (fake stationary regime) for inhomogeneous affine Stochastic Volterra equations.
result Existence of limiting distributions in the long run, which may depend on initial state.
Paper designs a bandit algorithm without reward distribution info.
problem Designing bandit algorithms without reward distribution info.
method Alternates between greedy rule and forced exploration.
result Achieves substantial regret upper bounds.
Reservoir Computing with Error Correction: Long-term Behaviors of Stochastic Dynamical Systemsmath.DS
Proposes a new method combining Reservoir Computing and Normalizing Flow for predicting stochastic dynamical systems.
problem Predicting and capturing long-term behaviors of stochastic dynamical systems.
method Data-driven framework combining Reservoir Computing and Normalizing Flow, integrating error modeling and both approaches virtues.
result Successfully predicts the long-term evolution of stochastic dynamical systems and replicates dynamical behaviors.
We find stationary distributions in a financial model with trends and mean-reversion.
problem Financial markets with competing trends and mean-reversion.
method Analytical derivation of stationary distributions in various noise and feedback regimes.
result The distributions are unimodal Gaussians in small noise, small feedback limits, but can be bimodal for stronger trends.
Study shows singularity of stationary measure on Furstenberg boundary for certain random walks.
problem Singularity of stationary measure on Furstenberg boundary for random walks.
method Analysis of random walks on semisimple Lie groups with specific properties.
result Stationary measure is singular to Lebesgue measure in certain cases.
Generating realistic asset-class scenarios from time series and curves
problem Simulating realistic trajectories for asset classes
method Combining parametric and resampling techniques
result More coherent and realistic simulations of yield-curve dynamics
Gradient clipping helps private SGD converge despite potential bias.
problem Gradient clipping in private SGD can bias convergence.
method Theoretical analysis and empirical evaluation of gradient clipping effects.
result Gradient clipping can prevent convergence to stationary points and introduces bias.