Regular Steps Research Articles

A Unified Walking Model for Dimeric Motor Proteins

Dimeric motor proteins, kinesin-1, cytoplasmic dynein-1, and myosin-V, move stepwise along microtubules and actin filaments with a regular step size. The motors take backward as well as forward steps. The step ratio r and dwell time τ, which are the ratio of the number of backward steps to the number of forward steps and the time between consecutive steps, respectively, were observed to change with the load. To understand the movement of motor proteins, we constructed a unified and simple mathematical model to explain the load dependencies of r and of τ measured for the above three types of motors quantitatively. Our model consists of three states, and the forward and backward steps are represented by the cycles of transitions visiting different pairs of states among the three, implying that a backward step is not the reversal of a forward step. Each of r and τ is given by a simple expression containing two exponential functions. The experimental data for r and τ for dynein available in the literature are not sufficient for a quantitative analysis, which is in contrast to those for kinesin and myosin-V. We reanalyze the data to obtain r and τ of native dynein to make up the insufficient data to fit them to the model. Our model successfully describes the behavior of r and τ for all of the motors in a wide range of loads from large assisting loads to superstall loads.

OBJECT BOUNDARIES REGULARIZATION USING THE DYNAMIC POLYLINE COMPRESSION ALGORITHM

Abstract. This study presents a regularization approach to refine object boundaries for the purpose of buildings 3D modelling and reconstruction. Specifically, the derivative Normalized Digital Surface model (nDSM) image layer is firstly segmented using the classical multi-resolution segmentation followed by spectral difference segmentation. As the segmentation results can contain quite a number of boundary artefacts in the form geometrical distortions, the Dynamic Polyline Compression algorithm (DCPA) is applied as a regularization step in order to refine the outer boundaries, which removes the distortions. This results in higher quality image objects for the purpose of 3D models reconstruction. Experimental results after comparing between automatically extracted buildings and manually digitized aerial photographs indicate high completeness scores of 94%–97% and correctness of 93%–96%. Overall average error is minimized with very low Root Mean Square (RMS) and Overlay errors.

Sparse kernel deep stacking networks

This article introduces the supervised deep learning method sparse kernel deep stacking networks (SKDSNs), which extend traditional kernel deep stacking networks (KDSNs) by incorporating a set of data-driven regularization and variable selection steps to improve predictive performance in high-dimensional settings. Before model fitting, variable pre-selection is carried out using genetic algorithms in combination with the randomized dependence coefficient, accounting for non-linear dependencies among the inputs and the outcome variable. During model fitting, internal variable selection is based on a ranked feature ordering which is tuned within the model-based optimization framework. Further regularization steps include $$L_1$$ -penalized kernel regression and dropout. Our simulation studies demonstrate an improved prediction accuracy of SKDSNs compared to traditional KDSNs. Runtime analysis of SKDSNs shows that the dimension of the random Fourier transformation greatly affects computational efficiency, and that the speed of SKDSNs can be improved by applying a subsampling-based ensemble strategy. Numerical experiments show that the latter strategy further increases predictive performance. Application of SKDSNs to three biomedical data sets confirm the results of the simulation study. SKDSNs are implemented in a new version of the R package kernDeepStackNet.

Double-layer stepped Si(1 0 0) surfaces prepared in As-rich CVD ambience

Low-defect III-V integration on Si(1 0 0) requires atomically well-ordered heterointerfaces. To this end, we study the interaction of As with Si(1 0 0) surfaces and the formation of dimerized Si(1 0 0):As surfaces in situ with reflection anisotropy spectroscopy during metalorganic chemical vapor deposition (MOCVD). Control of surface processes allows us to intentionally ‘rotate’ the dimer orientation and choose between predominantly (1 × 2) and (2 × 1) reconstructed Si(1 0 0):As surfaces both for low and higher surface misorientation. We also verify that the (2 × 1) surface reconstruction becomes energetically more favorable for vicinal Si(1 0 0):As surfaces. We explain the preparation of Si(1 0 0):As surfaces displaying an almost single-domain, (1 × 2) reconstructed surface with an RMS roughness ≤1 Å and exhibiting broad, atomically flat terraces as well as regular double-layer steps. This surface is highly suitable for subsequent low-defect III-V integration.

Block Matching Local SVD Operator Based Sparsity and TV Regularization for Image Denoising

We propose a denoising method by integrating group sparsity and TV regularization based on self-similarity of the image blocks. By using the block matching technique, we introduce some local SVD operators to get a good sparsity representation for the groups of the image blocks. The sparsity regularization and TV are unified in a variational problem and each of the subproblems can be efficiently optimized by splitting schemes. The proposed algorithm mainly contains the following four steps: block matching, basis vectors updating, sparsity regularization and TV smoothing. The self-similarity information of the image is assembled by the block matching step. By concatenating all columns of the similar image block together, we get redundancy matrices whose column vectors are highly correlated and should have sparse coefficients after a proper transformation. In contrast with many transformation based denoising methods such as BM3D with fixed basis vectors, we update local basis vectors derived from the SVD to enforce the sparsity representation. This step is equivalent to a dictionary learning procedure. With the sparsity regularization step, one can remove the noise efficiently and keep the texture well. The TV regularization step can help us to reduced the artifacts caused by the image block stacking. Besides, we mathematically show the convergence of the algorithms when the proposed model is convex (with $$p=1$$ ) and the bases are fixed. This implies the iteration adopted in BM3D is converged, which was not mathematically shown in the BM3D method. Numerical experiments show that the proposed method is very competitive and outperforms state-of-the-art denoising methods such as BM3D.

Microtubule Buckling Affects Cargo Transportation By Molecular Motor

A cell can be best resembled by a metropolitan city where all its functions are driven by the interactions of densely packed proteins. For the traffic system in this ‘city’, the expressway consists of actin and microtubules (MTs). On the other hand, the stepping motor proteins serve as ‘vehicles’ and transport cargo from the nucleus (minus end) to the periphery (plus end) and vice versa [1]. Kinesin-1 and cytoplasmic dynein are the two major cytoplasmic motors that transport cargos towards the plus and minus end of the MT respectively. Stimulated or spontaneous cell contraction, or constrained MT polymerization at the cell periphery give rise to buckling deformation of MT in cells, that may result into buckling with short wavelengths (e.g., several µm) [2-5]. While significant advancement is made in understanding the role of kinesin in transporting cargo individually or as team along undeformed MT [6-8], how the deformity in the MT due to buckling may affect the motility of the motor is overlooked. Kinesin takes regular steps of 8 nm in a hand-over-hand fashion [9] whereas dynein shows variable step sizes ranging from 8 – 32 nm uncoordinatedly [10]. Advanced studies on the structure and mechanism still do not provide clear idea on how the uncoordinated stepping mechanism in dynein responds to the buckled MT. Therefore, the effect of buckling of MT on the cargo transportation has a large scope to explore. Previously, we developed an experimental set-up, micro-stretcher, to control buckling of the MT filaments on a 2D elastic medium [11]. Present study exploits the knowledge from the system of MT buckling to further explore its contribution on the dynein based cargo transportation in vitro. For this, we fixed MTs on a pre-stretched elastic substrate, polydimethylsiloxane (PDMS), through interaction with kinesins. Relaxation of the pre-stretched PDMS consequently developed axial compression stress on the MTs, resulting into their buckling. We controlled the release of tensile strain of the PDMS to tune the buckling extent of MTs. Next, we immobilized the MTs with mild glutaraldehyde treatment and allowed the transportation of Qdots by motors along the undeformed and deformed MTs. We analyzed different parameters, e.g. velocities, pauses and run lengths from the observation of the cargo transportation. With increasing MT buckling we observed changes in the cargo transportation behavior of the motors. For kinesins, we observed that, the cargo velocities decreased with increasing strain in the MTs and finally, cargo transportation was halted at high strain region. While for dynein, cargo transportation prevailed with increased overall velocity with two peaks, one occurring in the low and the other at high strain region. Our goal is to explore the mechanism of how the buckling deformation of MT is affecting the observed phenomena in motor protein-based cargo transportation. Structural deformation in MTs, particularly, have showed to implicate several neurodegenerative diseases such as Alzheimer's disease, cardiovascular syndromes, neurological disorders and cancers etc. [12]. Therefore in vitro study of motor proteins in transporting cargo along deformed MTs will contribute not only in understanding the mechanism of motor motility in cells but also to target specific pathways for various neurodegenerative diseases.

Golden ratio stack of spirals for flexible angiographic imaging: Proof of concept in congenital heart disease.

In this study, a golden ratio stack of spiral (GRASS) sequence that used both golden step and golden angle ordering was implemented. The aim was to demonstrate that GRASS acquisitions could be flexibly reconstructed as both cardiac-gated and time-resolved angiograms. Image quality of time-resolved and cardiac-gated reconstructions of the GRASS sequence were compared to 3 conventional stack of spirals (SoS) acquisitions in an in silico model. In 10 patients, the GRASS sequence was compared to conventional breath hold angiography (BH-MRA) in terms of image quality and for vessel measurement. Vessel measurements were also compared to cine images. In the cardiac-gated in silico model, the image quality of GRASS was superior to regular and golden-angle with regular step SoS approaches. In the time-resolved model, GRASS image quality was comparable to the golden-angle with regular step technique and superior to regular SoS acquisitions. In patients, there was no difference in qualitative image scores between GRASS and BH-MRA, but SNR was lower. There was good agreement in vessel measurements between the GRASS reconstructions and conventional MR techniques (BH-MRA: 29.8 ± 5.6 mm, time-resolved GRASS-MRA: 29.9 ± 5.4 mm, SSFP diastolic: 29.4 ± 5.8 mm, cardiac-gated GRASS-MRA diastolic: 29.5 ± 5.5 mm, P > 0.87). We have demonstrated that the GRASS acquisition enables flexible reconstruction of the same raw data as both time-resolved and cardiac-gated volumes. This may enable better interrogation of anatomy in congenital heart disease.

New advanced concept of the flying capacitor asymmetric multilevel voltage source inverter

PurposeTo improve the voltage quality in AC adjustable high-power-speed-drive applications, the purpose of the paper is to provide a large number of output levels without increasing the number of commutation cells in the three-phase, n-cells flying capacitor voltage source asymmetric Multilevel Inverter (MI). The concept is based on the selection of different ratios between the breakdown voltages of two successive power devices. The new mathematical model is developed under various ratios, allows a thorough investigation of the harmonic distortions, flying capacitor energy storage, flying capacitor voltage balancing controllability and blocking voltage insulated gate bipolar transistor (IGBT) capability.Design/methodology/approachThe asymmetrical design provides a large number of output levels without increasing the number of commutation cells. The important new analytical expression of capacitors voltage distribution is derived and extended to any ratio between the switch breakdown voltages of two successive power devices.FindingsThe detailed simulation study of the proposed concept has been carried out using MATLAB/Simulink. The power switches control of the three-phase three-cell MI is assured by new phase-shifted-multi-carrier pulse width modulation. The space vector representation is used to show the regular and irregular step output voltage in the complex plan (α,β).Originality/valueIn the paper, the n cells flying capacitor inverter, which typically operates in the (n + 1) levels mode, was extended to (n + 2), (n + 3) … until 2n levels with regular or irregular step output voltage. Consequently, the claimed advantages of the asymmetric MI are to improve power quality by reducing harmonic distortions and to reduce the requirement on capacitive energy storage in the circuit.

Regularized lattice Boltzmann model for immiscible two-phase flows with power-law rheology.

In this work, a regularized lattice Boltzmann color-gradient model is developed for the simulation of immiscible two-phase flows with power-law rheology. This model is as simple as the Bhatnagar-Gross-Krook (BGK) color-gradient model except that an additional regularization step is introduced prior to the collision step. In the regularization step, the pseudo-inverse method is adopted as an alternative solution for the nonequilibrium part of the total distribution function, and it can be easily extended to other discrete velocity models no matter whether a forcing term is considered or not. The obtained expressions for the nonequilibrium part are merely related to macroscopic variables and velocity gradients that can be evaluated locally. Several numerical examples, including the single-phase and two-phase layered power-law fluid flows between two parallel plates, and the droplet deformation and breakup in a simple shear flow, are conducted to test the capability and accuracy of the proposed color-gradient model. Results show that the present model is more stable and accurate than the BGK color-gradient model for power-law fluids with a wide range of power-law indices. Compared to its multiple-relaxation-time counterpart, the present model can increase the computing efficiency by around 15%, while keeping the same accuracy and stability. Also, the present model is found to be capable of reasonably predicting the critical capillary number of droplet breakup.

Kinetic Nearest Neighbor Search in Black-Box Model

Proximity problems is a class of important problems which involve estimation of distances between geometric objects. The nearest neighbor search which is a subset of proximity problems, arises in numerous fields of applications, including Pattern Recognition, Statistical classification, Computer vision and etc. In this study, a nearest neighbor search is presented to move points in the plane, while query point is static. The proposed method works in the black-box KDS model, in which the points location received at regular time steps while at the same time, an upper bound dmax is known on the maximum displacement of any point at each time step. In this paper, a new algorithm is presented for kinetic nearest neighbor search problem in the black-box model, under assumptions on the distribution of the moving point set P. It has been shown how the kinetic nearest neighbor will be updated at each time step in O(k∆k log n) amortized time, where ∆k is the k-spread of a point set P . Key words: Computational Geometry, Black Box Model, Kinetic, Nearest Neighbor.

Linear Intensity-Based Image Registration

The accurate detection and localization of lesion within the prostate could greatly benefit in the planning of surgery and radiation therapy. Although T2 Weighted Imaging (T2WI) Magnetic Resonance Imaging (MRI) provides an infinite amount of anatomical information, which ease and improve diagnosis and patient treatment, however, modality specific image artifacts, such as the occurrences of intensity inhomogeneity are still obvious and can adversely affect quantitative image analysis. Conventional high resolution T2WI has been restricted in this respect. On the contrary, Apparent Diffusion Coefficient (ADC) map has been seen as capable to tackle T2WI limitation when a functional assessment of the prostate capable to provide added value compared to T2WI alone. Likewise, it has been shown that diagnosis using ADC map combined with T2WI significantly outperforms T2WI alone. Therefore, to obtain high accuracy detection and localization, a combination of high-resolution anatomic and functional imaging is extremely important in clinical practice. This strategy relies on accurate intensity based image registration. However, registration of anatomical and functional MR imaging is really challenging due to missing correspondences and intensity inhomogeneity. To address this problem, this study researches the used of applying linear geometric transform to the corresponding point to accurately mapping the images for precise alignment and accurate detection. Transformation type is crucial for the success of image registration. The selection of transformation type is influenced by the type and severity of the geometric differences between corresponding images, the accuracy of the control point between images, its density and organization of the control points. A transformation type is selected to reflect geometric differences between two images in image registration. Often, the selection of the suitable transformation type for image registration is undeniably challenging. To make this selection as effective as possible, an experimental mechanism has to be carried out to determine its suitability. These transformations types are Affine, similarity, rigid and translation. Additionally, intensity based image registration is implemented to optimize the similarity metric mean square error through regular step gradient descent optimizer. Accuracies evaluation for each transformation type has been carried out through mean square error (MSE) and peak signal noise ratio (PSNR). The results have been presented in a chart form together with a comparison table.

SIS: An R Package for Sure Independence Screening in Ultrahigh-Dimensional Statistical Models

We revisit sure independence screening procedures for variable selection in generalized linear models and the Cox proportional hazards model. Through the publicly available R package SIS, we provide a unified environment to carry out variable selection using iterative sure independence screening (ISIS) and all of its variants. For the regularization steps in the ISIS recruiting process, available penalties include the LASSO, SCAD, and MCP while the implemented variants for the screening steps are sample splitting, data-driven thresholding, and combinations thereof. Performance of these feature selection techniques is investigated by means of real and simulated data sets, where we find considerable improvements in terms of model selection and computational time between our algorithms and traditional penalized pseudo-likelihood methods applied directly to the full set of covariates.

Atlas-Based 3D Intensity Volume Reconstruction from 2D Long Leg Standing X-Rays: Application to Hard and Soft Tissues in Lower Extremity.

In this chapter, the reconstruction of 3D intensity volumes of femur, tibia, and three muscles around the thigh region from a pair of calibrated X-ray images is addressed. We present an atlas-based 2D-3D intensity volume reconstruction approach by combining a 2D-2D nonrigid registration-based 3D landmark reconstruction procedure with an adaptive regularization step. More specifically, an atlas derived from the CT acquisition of a healthy lower extremity, together with the input calibrated X-ray images, is used to reconstruct those musculoskeletal structures. To avoid the potential penetration of the reconstructed femoral and tibial volumes that might be caused by reconstruction error, we come up with an articulated 2D-3D reconstruction strategy, which can effectively preserve knee joint structure. Another contribution from our work is the application of the proposed 2D-3D reconstruction pipeline to derive the patient-specific volumes of three thigh muscles around the thigh region.

Migration policies in dynamic island models

Dynamic island models are population-based algorithms for solving optimization problems, where the individuals of the population are distributed on islands. These subpopulations of individuals are processed by search algorithms on each island. In order to share information within this distributed search process, the individuals migrate from their initial island to another destination island at regular steps. In dynamic island models, the migration process evolves during the search according to the observed performance on the different islands. The purpose of this dynamic/adaptive management of the migrations is to send the individuals to the most promising islands, with regards to their current states. Therefore, our approach is related to the adaptive management of search operators in evolutionary algorithms. In this work, our main purpose is thus to precisely analyze dynamic migration policies. We propose a testing process that assigns gains to the algorithms applied on the islands in order to assess the adaptive ability of the migration policies, with regards to various scenarios. Instead of having one dynamic migration policy that is applied to the whole search process, as it has already been studied, we propose to associate a migration policy to each individual, which allows us to combine simultaneously different migration policies.

Lab Analyses will Help Determine Filter Media Conditions

Laboratory analysis used as a regular maintenance step can support filter media assessment.

Adsorption properties and preparation of porous TaC ceramics with regular steps

Porous TaC ceramics were successfully prepared by particle packing method in order to improve the high temperature adsorption properties of ceramics. A large number of circular steps can be found clearly on the surface of the porous ceramic, and TaC ceramic particles are polycrystalline structures (fcc structure of NaCl-type). Subsequently, a porous TaC ceramics growth model was proposed to explain the formation of the steps on the ceramic surface, which revealed that the main reason for this shape was the overflow of CO and the melt of Ta2O5. Additionally, the specific surface area and median pore diameter of porous TaC ceramics were 0.098 (m2/g) and 3129 nm, respectively. And the porosity of the porous TaC ceramics was as high as 65.321%. It was impressive that the adsorption effects of porous TaC ceramics are remarkable (dropped from 4244 NTU to 0.3 NTU) and recyclable.

Symbolic pattern recognition for sequential data

ABSTRACTSources of sequential data surround and pervade our lives. Our bodies continuously generate sequential data such as heart rate and blood pressure. In global finance, stock indices and currency exchange rates change every second. The movement of clouds, the coordinates of the planets, the score of a soccer game, etc., are all examples of sequential data transitioning from one state to another in regular time steps. There is a mature body of literature related to modeling time-dependent sequential data, or time series, in which every model relies upon specific assumptions for applicability to data. However, there other kinds of sequential data that are not collected with respect to time; for example, DNA sequences (if we ignore the gene mutations). Modeling of this type of sequential data does not have a body of literature that is as mature. This is somewhat perplexing, since sequential data modeling should cover both time-dependent and non-time-dependent data.In this study, we introduce a framework called Symbolic Pattern Recognition for modeling pattern transition behavior of sequential data that is expressed by a finite alphabet of symbols. Our framework can be used to characterize, predict, simulate, cluster, and classify multiple series based on their observed pattern transition behaviors.We document our proposed framework and apply it to perform unsupervised clustering of 13 different species of mollusks based on their DNA sequences. Our model correctly clusters the mollusks into their respective genera, matching results obtained via more traditional, supervised genetic analysis.

Recursive regularization step for high-order lattice Boltzmann methods.

A lattice Boltzmann method (LBM) with enhanced stability and accuracy is presented for various Hermite tensor-based lattice structures. The collision operator relies on a regularization step, which is here improved through a recursive computation of nonequilibrium Hermite polynomial coefficients. In addition to the reduced computational cost of this procedure with respect to the standard one, the recursive step allows to considerably enhance the stability and accuracy of the numerical scheme by properly filtering out second- (and higher-) order nonhydrodynamic contributions in under-resolved conditions. This is first shown in the isothermal case where the simulation of the doubly periodic shear layer is performed with a Reynolds number ranging from 10^{4} to 10^{6}, and where a thorough analysis of the case at Re=3×10^{4} is conducted. In the latter, results obtained using both regularization steps are compared against the Bhatnagar-Gross-Krook LBM for standard (D2Q9) and high-order (D2V17 and D2V37) lattice structures, confirming the tremendous increase of stability range of the proposed approach. Further comparisons on thermal and fully compressible flows, using the general extension of this procedure, are then conducted through the numerical simulation of Sod shock tubes with the D2V37 lattice. They confirm the stability increase induced by the recursive approach as compared with the standard one.

Musical stimulation with feedback in gait training for Parkinson’s disease.

Patients suffering from idiopathic Parkinson's disease (PD) are unable to steadily generate regular steps due to an impairment of the basal ganglia that do not cue the supplementary motor area properly (Iansek, Bradshaw, Phillips, Cunnington, & Morris, 1995). This results in the main hallmarks of parkinsonian gait: reduced step cadence, reduced velocity, and shortened stride length. In the 1990s the use of acoustic stimulation for the rehabilitation of gait disorders in PD was developed under the concept of rhythmic auditory stimulation (RAS). RAS consists of a clear accentuated, either metronomic or musical rhythm that stimulates the production of rhythmical movements (Thaut, 2005), especially with respect to their timing. The consistent coupling between steady tempo of the acoustic stimulation (bpm, beats per minute) and walking cadence (spm, steps per minute) of PD patients has This been systematically investigated and the effectiveness of RAS broadly attested (for reviews see Lim et al., 2005; Nombela, Hughes, Owen, & Grahn, 2013; Tomlinson et al., 2013).In the standard RAS protocol, stimulation tempo usually corresponds to baseline cadence increased by 10% (Arias & Cudeiro, 2008; Freedland et al., 2002; Hausdorff, Lownthal, Herman, Gruendlinger, Peretz, & Giladi, 2007; Willems et al., 2006). RAS exerts a direct effect upon cadence and a consequent effect upon stride length. Because stride length in PD patients is lower than in healthy people at any given velocity (Ebersbach et al., 1999), it may be beneficial for the patient to address it specifically, as an independent therapy goal.Our choice of using musical instead of metronomic stimuli follows music therapist Michael Thaut's approach, where rhythmic stimuli are embedded in a musical structure. The approach is based on findings that rhythmic patterns within a musical context reduce response variability and synchronization offset more effectively than do single-pulse pattern in the frequency range of 60 to 120 spm (Thaut et al., 1996). Moreover, it has been demonstrated that healthy subjects synchronize more precisely (i.e., with lower negative mean asynchrony) with musical rather than metronomic stimuli (Repp, 2005). In the early studies with PD patients, RAS consisted of instrumental pieces in 2/4 or 4/4 m with rhythmic on-beats enhanced by a metronome (McIntosh, Brown, Rice, & Thaut, 1997; Miller, Thaut, McIntosh, & Rice, 1996; Thaut et al., 1996), and it was suggested that, compared with metronome, musical texture provides additional timing information and may therefore facilitate detection, anticipation, and synchronization to the beat (Thaut, Rathbun, & Miller, 1997). Moreover, music facilitates emotional involvement and motivates engagement in a task, and might therefore be a more efficient stimulation than the plain metronome beat. Recently, attention to the comparison between metronomic and musical stimulation and to the specific features that would account for the effectiveness of music seems to be increasing (Buhmann, Desmet, Moens, Van Dyck, & Leman, 2016; Styns, van Noorden, Moelants, & Leman, 2007; Wittwer, Webster, & Hill, 2013). Moreover, the specific effectiveness of music-based movement therapy on PD has been investigated in a meta-analysis (de Dreu, van der Wilk, Poppe, Kwakkel, & van Wegen, 2012).The fundamental advantage of the RAS approach is that it is able to replace the impaired internal stimuli of a parkinsonian brain. However, as stimulation proceeds independently from movement execution, there is no feedback mechanism that helps evaluate movements' adequacy. The potential of concurrent sensory feedback for enhancing motor learning has been extensively studied on healthy subjects (Sigrist, Rauter, Riener, & Wolf, 2013) and an interactive, feedback-like version of RAS has been recently proposed for the training of stride time in PD patients (Hove1, Suzuki, Uchitomi, Orimo, & Miyake, 2012). …

Computing monotone policies for Markov decision processes: a nearly-isotonic penalty approach

This paper discusses algorithms for solving Markov decision processes (MDPs) that have monotone optimal policies. We propose a two-stage alternating convex optimization scheme that can accelerate the search for an optimal policy by exploiting the monotone property The first stage is a linear program formulated in terms of the joint state-action probabilities. The second stage is a regularized problem formulated in terms of the conditional probabilities of actions given states. The regularization uses techniques from nearly-isotonic regression. While a variety of iterative method can be used in the first formulation of the problem, we show in numerical simulations that, in particular, the alternating method of multipliers (ADMM) can be significantly accelerated using the regularization step.