The Minimum-Variance Theory Revisited

Jf Feng

DOI: https://doi.org/10.1007/3-540-44868-3_9

2003-01-01

Abstract:The minimum-variance theory [12] was proposed to account for the eye and arm movement. However, we point out here that i) the input signals used in the the simulations are not Poisson processes; ii) when the input signal is a Poisson process, the solution of the minimum-variance is degenerate.

What problem does this paper attempt to address?

Towards A Mathematical Foundation Of Minimum-Variance Theory

Jf Feng,Kw Zhang,G Wei

DOI: https://doi.org/10.1088/0305-4470/35/34/304

2002-01-01

Abstract:The minimum-variance theory which accounts for arm and eye movements with noise signal inputs was proposed by Harris and Wolpert (1998 Nature 394 780-4). Here we present a detailed theoretical analysis of the theory and analytical solutions of the theory are obtained. Furthermore, we propose a new version of the minimum-variance theory, which is more realistic for a biological system. For the new version we show numerically that the variance is considerably reduced.
A Note on Minimum-Variance Theory and Beyond

Jf Feng,G Tartaglia,B Tirozzi

DOI: https://doi.org/10.1088/0305-4470/37/17/001

2004-01-01

Abstract:We revisit the minimum-variance theory proposed by Harris and Wolpert (1998 Nature 394 780-4), discuss the implications of the theory on modelling the firing patterns of single neurons and analytically find the optimal control signals, trajectories and velocities. Under the rate coding assumption, input control signals employed in the minimum-variance theory should be Fitts processes rather than Poisson processes. Only if information is coded by interspike intervals, Poisson processes are in agreement with the inputs employed in the minimum-variance theory. For the integrate-and-fire model with Fitts process inputs, interspike intervals of efferent spike trains are very irregular. We introduce diffusion approximations to approximate neural models with renewal process inputs and present theoretical results on calculating moments of interspike intervals of the integrate-and-fire model. Results in Feng, et al (2002 J. Phys. A: Math. Gen. 35 7287-304) are generalized. In conclusion, we present a complete picture on the minimum-variance theory ranging from input control signals, to model outputs, and to its implications on modelling firing patterns of single neurons.
Minimum Contrast Estimator for Fractional Ornstein-Uhlenbeck Processes

WeiLin Xiao,WeiGuo Zhang,XiLi Zhang

DOI: https://doi.org/10.1007/s11425-012-4386-y

2012-01-01

Science China Mathematics

Abstract:This paper proposes a minimum contrast methodology to estimate the drift parameter for the Ornstein-Uhlenbeck process driven by fractional Brownian motion of Hurst index, which is greater than one half. Both the strong consistency and the asymptotic normality of this minimum contrast estimator are studied based on the Laplace transform. The numerical simulation results confirm the theoretical analysis and show that the minimum contrast technique is effective and efficient.
Minimum Variance Prediction of MIMO Linear Time-varying Systems

Zheng Li

DOI: https://doi.org/10.1007/s12555-024-0446-7

IF: 2.964

2024-10-06

International Journal of Control Automation and Systems

Abstract:In this paper we study the problem of minimum variance prediction of linear time-varying systems and develop a minimum variance predictor for a standard multi-input multi-output time-varying ARMA model. It is shown that the linear time-varying predictor allows time-varying degrees in the ARMA model and minimizes the output prediction error variance under colored noise and rapidly time-varying plant parameters. Assumptions made on the linear time-varying systems are natural extensions of those made on linear time-invariant plants by the standard linear time-invariant minimum variance predictor.

automation & control systems
Stochastic Variational Approach to Minimum Uncertainty States

F. Illuminati,L. Viola

DOI: https://doi.org/10.1088/0305-4470/28/10/022

1995-01-11

Abstract:We introduce a new variational characterization of Gaussian diffusion processes as minimum uncertainty states. We then define a variational method constrained by kinematics of diffusions and Schrödinger dynamics to seek states of local minimum uncertainty for general non-harmonic potentials.

High Energy Physics - Theory,Condensed Matter,Quantum Physics
Maximum Principle for Mean Field Type Control Problems with General Volatility Functions

Alain Bensoussan,Ziyu Huang,Sheung Chi Phillip Yam

2023-09-13

Abstract:In this paper, we study the maximum principle of mean field type control problems when the volatility function depends on the state and its measure and also the control, by using our recently developed method. Our method is to embed the mean field type control problem into a Hilbert space to bypass the evolution in the Wasserstein space. We here give a necessary condition and a sufficient condition for these control problems in Hilbert spaces, and we also derive a system of forward-backward stochastic differential equations.

Optimization and Control
Maximum Principle for Mean-Field SDEs Under Model Uncertainty

Wei He,Peng Luo,Falei Wang

DOI: https://doi.org/10.1007/s00245-023-09970-8

2023-01-01

Abstract:This paper is concerned with necessary and sufficient conditions for optimal control problem of mean-field type under model uncertainty. Since the classical variational method is not enough to get the results in this case, we should draw support from weak convergence method. As an application in finance, we study the time-inconsistent mean-variance portfolio selection problem in this model uncertainty setup.
Three Laws of Multivariable Feedback Systems, Extended Spectral Flatness (Extended Wiener Entropy), 'Uncertainty Principles' in Variance Minimization, and Performance Limitations in Minimum Variance Estimation/Filtering

Song Fang

DOI: https://doi.org/10.48550/arXiv.1412.1056

2014-12-19

Abstract:In this paper, three laws are obtained for multiple-input multiple-output feedback systems, which are in entropy domain, frequency domain, and time domain, respectively. The system setup is that with causal plants and causal controllers. Those laws characterize the performance limitations of such systems imposed by the feedback mechanism. Some new notions are proposed to facilitate the analysis: negentropy rate, extended spectral flatness (extended Wiener entropy), Gaussianity-whiteness measure (joint Shannon-Wiener entropy), etc. Two approaches are adopted: the integrated approach and the divided approach. And 'uncertainty principles' are found in minimum variance control. Besides, performance limitations in minimum variance estimation and filtering are obtained. In the end, the special case of linear time-invariant feedback systems is discussed.

Systems and Control,Information Theory,Optimization and Control
Statistical inference for rough volatility: Minimax Theory

Carsten Chong,Marc Hoffmann,Yanghui Liu,Mathieu Rosenbaum,Grégoire Szymanski

2024-02-16

Abstract:Rough volatility models have gained considerable interest in the quantitative finance community in recent years. In this paradigm, the volatility of the asset price is driven by a fractional Brownian motion with a small value for the Hurst parameter $H$. In this work, we provide a rigorous statistical analysis of these models. To do so, we establish minimax lower bounds for parameter estimation and design procedures based on wavelets attaining them. We notably obtain an optimal speed of convergence of $n^{-1/(4H+2)}$ for estimating $H$ based on n sampled data, extending results known only for the easier case $H>1/2$ so far. We therefore establish that the parameters of rough volatility models can be inferred with optimal accuracy in all regimes.

Statistics Theory,Statistical Finance
Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements

Shay Ben-Itzhak,Amir Karniel

DOI: https://doi.org/10.1162/neco.2007.12-05-077

Abstract:Rapid arm-reaching movements serve as an excellent test bed for any theory about trajectory formation. How are these movements planned? A minimum acceleration criterion has been examined in the past, and the solution obtained, based on the Euler-Poisson equation, failed to predict that the hand would begin and end the movement at rest (i.e., with zero acceleration). Therefore, this criterion was rejected in favor of the minimum jerk, which was proved to be successful in describing many features of human movements. This letter follows an alternative approach and solves the minimum acceleration problem with constraints using Pontryagin's minimum principle. We use the minimum principle to obtain minimum acceleration trajectories and use the jerk as a control signal. In order to find a solution that does not include nonphysiological impulse functions, constraints on the maximum and minimum jerk values are assumed. The analytical solution provides a three-phase piecewise constant jerk signal (bang-bang control) where the magnitude of the jerk and the two switching times depend on the magnitude of the maximum and minimum available jerk values. This result fits the observed trajectories of reaching movements and takes into account both the extrinsic coordinates and the muscle limitations in a single framework. The minimum acceleration with constraints principle is discussed as a unifying approach for many observations about the neural control of movements.
Consistance d'un estimateur de minimum de variance étendue

Joseph Rynkiewicz

DOI: https://doi.org/10.48550/arXiv.0802.3190

2008-02-22

Abstract:We consider a generalization of the criterion minimized by the K-means algorithm, where a neighborhood structure is used in the calculus of the variance. Such tool is used, for example with Kohonen maps, to measure the quality of the quantification preserving the neighborhood relationships. If we assume that the parameter vector is in a compact Euclidean space and all it components are separated by a minimal distance, we show the strong consistency of the set of parameters almost realizing the minimum of the empirical extended variance.

Statistics Theory
Linear Variance Bounds for Particle Approximations of Time-Homogeneous Feynman-Kac Formulae

Nick Whiteley,Nikolas Kantas,Ajay Jasra

DOI: https://doi.org/10.1016/j.spa.2012.02.002

2012-02-13

Abstract:This article establishes sufficient conditions for a linear-in-time bound on the non-asymptotic variance of particle approximations of time-homogeneous Feynman-Kac formulae. These formulae appear in a wide variety of applications including option pricing in finance and risk sensitive control in engineering. In direct Monte Carlo approximation of these formulae, the non-asymptotic variance typically increases at an exponential rate in the time parameter. It is shown that a linear bound holds when a non-negative kernel, defined by the logarithmic potential function and Markov kernel which specify the Feynman-Kac model, satisfies a type of multiplicative drift condition and other regularity assumptions. Examples illustrate that these conditions are general and flexible enough to accommodate two rather extreme cases, which can occur in the context of a non-compact state space: 1) when the potential function is bounded above, not bounded below and the Markov kernel is not ergodic; and 2) when the potential function is not bounded above, but the Markov kernel itself satisfies a multiplicative drift condition.

Computation,Statistics Theory
Minimal Kullback–Leibler Divergence for Constrained Lévy–Itô Processes

Sebastian Jaimungal,Silvana M. Pesenti,Leandro Sánchez-Betancourt

DOI: https://doi.org/10.1137/23m1555697

IF: 2.2

2024-03-19

SIAM Journal on Control and Optimization

Abstract:SIAM Journal on Control and Optimization, Volume 62, Issue 2, Page 982-1005, April 2024. Given an [math]-dimensional stochastic process [math] driven by [math]-Brownian motions and Poisson random measures, we search for a probability measure [math], with minimal relative entropy to [math], such that the [math]-expectations of some terminal and running costs are constrained. We prove existence and uniqueness of the optimal probability measure, derive the explicit form of the measure change, and characterize the optimal drift and compensator adjustments under the optimal measure. We provide an analytical solution for Value-at-Risk (quantile) constraints, discuss how to perturb a Brownian motion to have arbitrary variance, and show that pinned measures arise as a limiting case of optimal measures. The results are illustrated in a risk management setting—including an algorithm to simulate under the optimal measure—and explore an example where an agent seeks to answer the question what dynamics are induced by a perturbation of the Value-at-Risk and the average time spent below a barrier on the reference process?

mathematics, applied,automation & control systems
On Minimum Trace Factor Analysis -- An Old Song Sung to a New Tune

C. Li,A. Shkolnik

2024-02-04

Abstract:Dimensionality reduction methods, such as principal component analysis (PCA) and factor analysis, are central to many problems in data science. There are, however, serious and well-understood challenges to finding robust low dimensional approximations for data with significant heteroskedastic noise. This paper introduces a relaxed version of Minimum Trace Factor Analysis (MTFA), a convex optimization method with roots dating back to the work of Ledermann in 1940. This relaxation is particularly effective at not overfitting to heteroskedastic perturbations and addresses the commonly cited Heywood cases in factor analysis and the recently identified "curse of ill-conditioning" for existing spectral methods. We provide theoretical guarantees on the accuracy of the resulting low rank subspace and the convergence rate of the proposed algorithm to compute that matrix. We develop a number of interesting connections to existing methods, including HeteroPCA, Lasso, and Soft-Impute, to fill an important gap in the already large literature on low rank matrix estimation. Numerical experiments benchmark our results against several recent proposals for dealing with heteroskedastic noise.

Machine Learning,Methodology
Minimum Information Variability in Linear Langevin Systems via Model Predictive Control

Adrian-Josue Guel-Cortez,Eun-jin Kim,Mohamed W. Mehrez

DOI: https://doi.org/10.3390/e26040323

IF: 2.738

2024-04-11

Entropy

Abstract:Controlling the time evolution of a probability distribution that describes the dynamics of a given complex system is a challenging problem. Achieving success in this endeavour will benefit multiple practical scenarios, e.g., controlling mesoscopic systems. Here, we propose a control approach blending the model predictive control technique with insights from information geometry theory. Focusing on linear Langevin systems, we use model predictive control online optimisation capabilities to determine the system inputs that minimise deviations from the geodesic of the information length over time, ensuring dynamics with minimum "geometric information variability". We validate our methodology through numerical experimentation on the Ornstein–Uhlenbeck process and Kramers equation, demonstrating its feasibility. Furthermore, in the context of the Ornstein–Uhlenbeck process, we analyse the impact on the entropy production and entropy rate, providing a physical understanding of the effects of minimum information variability control.

physics, multidisciplinary
Empirical Variance Minimization with Applications in Variance Reduction and Optimal Control

D. Belomestny,L. Iosipoi,Q. Paris,N. Zhivotovskiy

DOI: https://doi.org/10.48550/arXiv.1712.04667

2021-08-01

Abstract:We study the problem of empirical minimization for variance-type functionals over functional classes. Sharp non-asymptotic bounds for the excess variance are derived under mild conditions. In particular, it is shown that under some restrictions imposed on the functional class fast convergence rates can be achieved including the optimal non-parametric rates for expressive classes in the non-Donsker regime under some additional assumptions. Our main applications include variance reduction and optimal control.

Numerical Analysis,Machine Learning
Action Minimizing Stochastic Invariant Measures For A Class Of Lagrangian Systems

Kaizhi Wang

DOI: https://doi.org/10.3934/cpaa.2008.7.1211

2008-01-01

Abstract:In this paper we discuss a variational method of constructing an action minimizing stochastic invariant measure for positive definite Lagrangian systems. Then we study some main properties of the stochastic minimal measures. Finally we give the definitions of stochastic Mather's functions with respect to the stochastic differential equation dx = nu(t)dt + sigma(x)dw and prove their differentiability.
Modeling trade-off between time-optimal and minimum energy in saccade main sequence

Xuezhong Wang,Simon M Hsiang

DOI: https://doi.org/10.1007/s00422-011-0420-3

Abstract:Saccadic eye movement is highly stereotyped and commonly believed to be governed by an open-loop control mechanism. We propose a principle combining time-optimal and minimum control energy criteria to account for the saccade main sequence as observed from empirical data. The model prediction revealed that the weighting factor of the energy conservation becomes more dominant than the time-optimal when the saccade amplitude is large. We demonstrate that the proposed model is a general form synthesizing the time-optimum, minimum torque change, and minimum control effort models. In addition, we show the connection between our model and the stochastic minimum variance models from the aspect of optimization.
Model Misfit Minimization

Yuanyuan Fang,Ying Zhou,Zhenxing Yao

DOI: https://doi.org/10.1785/0120190079

2019-01-01

Bulletin of the Seismological Society of America

Abstract:In geophysical applications, solutions to ill-posed inverse problems Ax = b are often obtained by analyzing the trade-off between data residue parallel to Ax - b parallel to(2) and model norm parallel to x parallel to(2). In this study, we show that the traditional L-curve analysis does not lead to solutions closest to the true models because the maximum curvature (or the corner of the L-curve) depends on the relative scaling between data residue and model norm. A Bayes approach based on empirical risk function minimization using training datasets may be designed to find a statistically optimal solution, but its success depends on the true realization of the model. To overcome this limitation, we construct training models using eigenvectors of matrix A(T)A as well as spectral coefficients calculated from the correlation between observations and eigenvector projected data. This approach accounts for data noise level but does not require it as a priori knowledge. Using global tomography as an example, we show that the solutions are closest to true models.

The Minimum-Variance Theory Revisited

Towards A Mathematical Foundation Of Minimum-Variance Theory

A Note on Minimum-Variance Theory and Beyond

Minimum Contrast Estimator for Fractional Ornstein-Uhlenbeck Processes

Minimum Variance Prediction of MIMO Linear Time-varying Systems

Stochastic Variational Approach to Minimum Uncertainty States

Maximum Principle for Mean Field Type Control Problems with General Volatility Functions

Maximum Principle for Mean-Field SDEs Under Model Uncertainty

Three Laws of Multivariable Feedback Systems, Extended Spectral Flatness (Extended Wiener Entropy), 'Uncertainty Principles' in Variance Minimization, and Performance Limitations in Minimum Variance Estimation/Filtering

Statistical inference for rough volatility: Minimax Theory

Minimum acceleration criterion with constraints implies bang-bang control as an underlying principle for optimal trajectories of arm reaching movements

Consistance d'un estimateur de minimum de variance étendue

Linear Variance Bounds for Particle Approximations of Time-Homogeneous Feynman-Kac Formulae

Minimal Kullback–Leibler Divergence for Constrained Lévy–Itô Processes

On Minimum Trace Factor Analysis -- An Old Song Sung to a New Tune

Minimum Information Variability in Linear Langevin Systems via Model Predictive Control

Empirical Variance Minimization with Applications in Variance Reduction and Optimal Control

Action Minimizing Stochastic Invariant Measures For A Class Of Lagrangian Systems

Modeling trade-off between time-optimal and minimum energy in saccade main sequence

Model Misfit Minimization