arXiv research

The method approximates stationary distributions of Markov models by truncating irrelevant states.

New definitions of ESP for quantum reservoir computing handle non-stationary systems.

Study identifies transitions between traffic modes on Cologne motorways.

ETGPSSM efficiently models high-dimensional, non-stationary systems with reduced complexity.

Study neural architectures on learned latent graphs using Schrödinger dynamics.

Study measures rigidity for random walks and flows via generalized u-Gibbs states.

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 …

Non-stationary reinforcement learning is challenging due to the complexity of updating value functions.

New method optimizes policies in non-stationary environments.

Proposes a new algorithm for non-stationary bandits.

Study on fake stationary Volterra Heston model for non-stationary processes.

NVMDP framework tackles non-stationary MDPs with varying discount rates.

Using the formalism of Lyapunov potential function it is shown that the stability principles for biomass in the ecosystem and for employment in economics are mathematically similar. The ecosystem is found to have a stable and an unstable stationary state with high (forest) and low (grasslands) biomass, respectively. In…

KeRNS tackles non-stationary reinforcement learning in metric spaces.

DAL enhances black-box bandit algorithms for non-stationary environments.

Detect changes in noisy dynamical systems using empirical approximations and finite-sample bounds.

Enhances time-series regression trees with latent factors for robust financial analysis.

Proposes a method to forecast non-stationary time series.

MetaCURL tackles non-stationary MDPs with optimal dynamic regret.

Stanza models complex time series with balance between traditional and deep learning approaches.

Model predicts stationary equilibrium in investment decisions of firms in fluctuating markets.

We characterize stationary solutions to McKean-Vlasov equations on the circle.

DARLING tackles non-stationary RL with guarantees, improving dynamic regret.

Improved algorithm detects changes in RL environments with non-stationary MDPs.

We study how resetting affects geometric Brownian motion, showing it becomes stationary but remains non-ergodic.

Algorithm minimizes regret in non-stationary dueling bandits with unknown parameters.

Develops a new method to discover causal relationships from nonstationary time series data.

New Hida-Matérn kernels enable flexible process priors and efficient GP inference.

Develops RL algorithm for non-Markovian, non-stationary reward streams.

We combine geometric data analysis and stochastic modeling to describe the collective dynamics of complex systems. As an example we apply this approach to financial data and focus on the non-stationarity of the market correlation structure. We identify the dominating variable and extract its explicit stochastic model. …

In this paper, we consider a simple kinetic model of economy involving both exchanges between agents and speculative trading. We show that the kinetic model admits non trivial quasi-stationary states with power law tails of Pareto type. In order to do this we consider a suitable asymptotic limit of the model yielding a…

New algorithm tackles non-stationary reinforcement learning with general function approximation.

The state space (SS) representation of Gaussian processes (GP) has recently gained a lot of interest. The main reason is that it allows to compute GPs based inferences in O(n), where $n$ is the number of observations. This implementation makes GPs suitable for Big Data. For this reason, it is important to provide a SS …

Master algorithm fails to detect non-stationarity in practical settings.

Improved sampling from mean-field stationary distributions.

Learn true model from metastable samples of discrete distributions.

New method finds stationary points in bilevel optimization problems.

Building on previous results on the quadratic helicity in magnetohydrodynamics (MHD) we investigate particular minimum helicity states. Those are eigenfunctions of the curl operator and are shown to constitute solutions of the quasi-stationary incompressible ideal MHD equations. We then show that these states have inde…

The understanding of complex systems has become a central issue because complex systems exist in a wide range of scientific disciplines. Time series are typical experimental results we have about complex systems. In the analysis of such time series, stationary situations have been extensively studied and correlations h…

Develops a deep non-stationary kernel for non-stationary spatio-temporal point processes.

Natural spatiotemporal processes can be highly non-stationary in many ways, e.g. the low-level non-stationarity such as spatial correlations or temporal dependencies of local pixel values; and the high-level variations such as the accumulation, deformation or dissipation of radar echoes in precipitation forecasting. Fr…

New algorithms reduce regret in non-stationary bandits with increasing payoffs.

The notion of the stationary equilibrium ensemble has played a central role in statistical mechanics. In machine learning as well, training serves as generalized equilibration that drives the probability distribution of model parameters toward stationarity. Here, we derive stationary fluctuation-dissipation relations t…

New method stabilizes FQE by reweighting Bellman targets.

New approach turns optimal stationary RL into non-stationary RL without prior knowledge.

In this paper we study several aspects of the geometry of conformally stationary Lorentz manifolds, and particularly of GRW spaces, due to the presence of a closed conformal vector field. More precisely, we begin by extending to these spaces a result of J. Simons on the minimality of cones in Euclidean space, and apply…

PyChEst detects changes in non-stationary time series without distributional assumptions.

New algorithm tackles non-stationary RL with near-optimal regret bounds.