arXiv research

Framework infers Langevin dynamics from stochastic observations of latent systems.

New approach handles stochastic and partially-observable environments using discrete autoencoders and Monte Carlo tree search.

Improved averaging method for noisy observations converges strongly.

We study online convex optimization under stochastic sub-gradient observation faults, where we introduce adaptive algorithms with minimax optimal regret guarantees. We specifically study scenarios where our sub-gradient observations can be noisy or even completely missing in a stochastic manner. To this end, we propose…

Method learns dynamics from noisy partial observations.

Stochastic methods improve data assimilation with high-frequency sensor data.

New method learns stochastic process representations without exact reconstruction.

VSDN models sporadic time series with neural SDEs.

We extend probabilistic programming to handle conditioning on marginal distributions.

Paper improves neural ODEs for forecasting non-Markovian processes.

In reinforcement learning episodes, the rewards and punishments are often non-deterministic, and there are invariably stochastic elements governing the underlying situation. Such stochastic elements are often numerous and cannot be known in advance, and they have a tendency to obscure the underlying rewards and punishm…

Method learns dynamics of slow variables from stochastic data.

Paper develops efficient Bayesian inference for enzymatic SRNs with LNA metamodel.

Study examines adversarial robustness of ANN variants, revealing differences in black-box settings.

New model captures state-dependent variability in partially observed systems.

This paper studies the question of filtering and maximizing terminal wealth from expected utility in a partially information stochastic volatility models. The special features is that the only information available to the investor is the one generated by the asset prices, and the unobservable processes will be modeled …

New controller reduces regret in non-stochastic control with adversarial perturbations.

As the size of modern data sets exceeds the disk and memory capacities of a single computer, machine learning practitioners have resorted to parallel and distributed computing. Given that optimization is one of the pillars of machine learning and predictive modeling, distributed optimization methods have recently garne…

In this paper we present a novel quasi-Newton algorithm for use in stochastic optimisation. Quasi-Newton methods have had an enormous impact on deterministic optimisation problems because they afford rapid convergence and computationally attractive algorithms. In essence, this is achieved by learning the second-order (…

Framework learns stochastic dynamics from endpoint and intermediate distributions using soft energy constraints.

We study the exponential Ornstein-Uhlenbeck stochastic volatility model and observe that the model shows a multiscale behavior in the volatility autocorrelation. It also exhibits a leverage correlation and a probability profile for the stationary volatility which are consistent with market observations. All these featu…

A new Bayesian filtering method speeds up stochastic Newton optimization.

We present a framework for describing the evolution of stochastic observables having a non-stationary distribution of values. The framework is applied to empirical volume-prices from assets traded at the New York stock exchange. Using Kullback-Leibler divergence we evaluate the best model out from four biparametric mod…

We establish a stochastic maximum principle (SMP) for control problems of partially observed diffusions of mean-field type with risk-sensitive performance functionals.

SCOTCH learns system structure from irregular time series using neural SDEs.

IDS optimizes regret in stochastic partial monitoring with linear rewards.

Earlier we proposed the stochastic point process model, which reproduces a variety of self-affine time series exhibiting power spectral density S(f) scaling as power of the frequency f and derived a stochastic differential equation with the same long range memory properties. Here we present a stochastic differential eq…

This paper contains a summary of mathematical researches of stochastic properties of the long time behavior of a continuously observed (and interactively controlled) quantum--field top. Applications to interactively controlled stochastic computer-graphic dynamical systems are also discussed.

Paper develops a new algorithm for sparse signal recovery.

New estimator learns symmetric dynamics from few observations.

This paper provides a unifying theoretical framework for stochastic optimization algorithms by means of a latent stochastic variational problem. Using techniques from stochastic control, the solution to the variational problem is shown to be equivalent to that of a Forward Backward Stochastic Differential Equation (FBS…

Computational aspects of the optimal consumption and investment with the partially observed stochastic volatility of the asset prices are considered. The new quantization approach to filtering - density quantization - is introduced which reduces the original infinite dimensional state space of the problem to the finite…

Thompson Sampling tackles noisy context in stochastic bandits.

A new transformer model uses Monte Carlo methods for sequence prediction.

New algorithm achieves optimal regret in non-stochastic control, showing stochasticity is not beneficial.

We propose a method to infer stochastic low-rank RNNs from neural data.

We study the computational tractability of PAC reinforcement learning with rich observations. We present new provably sample-efficient algorithms for environments with deterministic hidden state dynamics and stochastic rich observations. These methods operate in an oracle model of computation -- accessing policy and va…

Paper extends a method to estimate Hurst parameter for rough stochastic volatility models.

New method for sampling from complex distributions using stochastic localization.

Model optimizes trading strategy with unobservable toxicity.

New model estimates higher-order interactions in stochastic processes using lower-dimensional projections.

New method learns dynamics from sparse data using geometric constraints.

This paper studies the complexity of the stochastic gradient algorithm for PCA when the data are observed in a streaming setting. We also propose an online approach for selecting the learning rate. Simulation experiments confirm the practical relevance of the plain stochastic gradient approach and that drastic improvem…

Develops anytime-valid stopping rules for SGD based on observed trajectory.

Paper addresses privacy and robustness in stochastic linear bandits.

Deep learning scheme identifies and reconstructs chaotic and stochastic systems from noisy data.

Motivated by the scarcity of accurate payoff feedback in practical applications of game theory, we examine a class of learning dynamics where players adjust their choices based on past payoff observations that are subject to noise and random disturbances. First, in the single-player case (corresponding to an agent tryi…

Deep density methods improve filtering in high-dimensional systems.