Zhishan Guo

A one-layer recurrent neural network for constrained pseudoconvex optimization and its application for dynamic portfolio optimization

In this paper, a one-layer recurrent neural network is proposed for solving pseudoconvex optimization problems subject to linear equality and bound constraints. Compared with the existing neural networks for optimization (e.g., the projection neural networks), the proposed neural network is capable of solving more general pseudoconvex optimization problems with equality and bound constraints. Moreover, it is capable of solving constrained fractional programming problems as a special case. The convergence of the state variables of the proposed neural network to achieve solution optimality is guaranteed as long as the designed parameters in the model are larger than the derived lower bounds. Numerical examples with simulation results illustrate the effectiveness and characteristics of the proposed neural network. In addition, an application for dynamic portfolio optimization is discussed.

Analyses of allele-specific gene expression in highly divergent mouse crosses identifies pervasive allelic imbalance

Complex human traits are influenced by variation in regulatory DNA through mechanisms that are not fully understood. Because regulatory elements are conserved between humans and mice, a thorough annotation of cis regulatory variants in mice could aid in further characterizing these mechanisms. Here we provide a detailed portrait of mouse gene expression across multiple tissues in a three-way diallel. Greater than 80% of mouse genes have cis regulatory variation. Effects from these variants influence complex traits and usually extend to the human ortholog. Further, we estimate that at least one in every thousand SNPs creates a cis regulatory effect. We also observe two types of parent-of-origin effects, including classical imprinting and a new global allelic imbalance in expression favoring the paternal allele. We conclude that, as with humans, pervasive regulatory variation influences complex genetic traits in mice and provide a new resource toward understanding the genetic control of transcription in mammals.

A One-Layer Recurrent Neural Network for Pseudoconvex Optimization Subject to Linear Equality Constraints

In this paper, a one-layer recurrent neural network is presented for solving pseudoconvex optimization problems subject to linear equality constraints. The global convergence of the neural network can be guaranteed even though the objective function is pseudoconvex. The finite-time state convergence to the feasible region defined by the equality constraints is also proved. In addition, global exponential convergence is proved when the objective function is strongly pseudoconvex on the feasible region. Simulation results on illustrative examples and application on chemical process data reconciliation are provided to demonstrate the effectiveness and characteristics of the neural network.

Efficient gene-driven germ-line point mutagenesis of C57BL/6J mice

BackgroundAnalysis of an allelic series of point mutations in a gene, generated by N-ethyl-N-nitrosourea (ENU) mutagenesis, is a valuable method for discovering the full scope of its biological function. Here we present an efficient gene-driven approach for identifying ENU-induced point mutations in any gene in C57BL/6J mice. The advantage of such an approach is that it allows one to select any gene of interest in the mouse genome and to go directly from DNA sequence to mutant mice.ResultsWe produced the Cryopreserved Mutant Mouse Bank (CMMB), which is an archive of DNA, cDNA, tissues, and sperm from 4,000 G1 male offspring of ENU-treated C57BL/6J males mated to untreated C57BL/6J females. Each mouse in the CMMB carries a large number of random heterozygous point mutations throughout the genome. High-throughput Temperature Gradient Capillary Electrophoresis (TGCE) was employed to perform a 32-Mbp sequence-driven screen for mutations in 38 PCR amplicons from 11 genes in DNA and/or cDNA from the CMMB mice. DNA sequence analysis of heteroduplex-forming amplicons identified by TGCE revealed 22 mutations in 10 genes for an overall mutation frequency of 1 in 1.45 Mbp. All 22 mutations are single base pair substitutions, and nine of them (41%) result in nonconservative amino acid substitutions. Intracytoplasmic sperm injection (ICSI) of cryopreserved spermatozoa into B6D2F1 or C57BL/6J ova was used to recover mutant mice for nine of the mutations to date.ConclusionsThe inbred C57BL/6J CMMB, together with TGCE mutation screening and ICSI for the recovery of mutant mice, represents a valuable gene-driven approach for the functional annotation of the mammalian genome and for the generation of mouse models of human genetic diseases. The ability of ENU to induce mutations that cause various types of changes in proteins will provide additional insights into the functions of mammalian proteins that may not be detectable by knockout mutations.

Flexible and robust co-regularized multi-domain graph clustering

Multi-view graph clustering aims to enhance clustering performance by integrating heterogeneous information collected in different domains. Each domain provides a different view of the data instances. Leveraging cross-domain information has been demonstrated an effective way to achieve better clustering results. Despite the previous success, existing multi-view graph clustering methods usually assume that different views are available for the same set of instances. Thus instances in different domains can be treated as having strict one-to-one relationship. In many real-life applications, however, data instances in one domain may correspond to multiple instances in another domain. Moreover, relationships between instances in different domains may be associated with weights based on prior (partial) knowledge. In this paper, we propose a flexible and robust framework, CGC (Co-regularized Graph Clustering), based on non-negative matrix factorization (NMF), to tackle these challenges. CGC has several advantages over the existing methods. First, it supports many-to-many cross-domain instance relationship. Second, it incorporates weight on cross-domain relationship. Third, it allows partial cross-domain mapping so that graphs in different domains may have different sizes. Finally, it provides users with the extent to which the cross-domain instance relationship violates the in-domain clustering structure, and thus enables users to re-evaluate the consistency of the relationship. Extensive experimental results on UCI benchmark data sets, newsgroup data sets and biological interaction networks demonstrate the effectiveness of our approach.

Metric Learning from Relative Comparisons by Minimizing Squared Residual

Recent studies [1] -- [5] have suggested using constraints in the form of relative distance comparisons to represent domain knowledge: d(a, b) <; d(c, d) where d(·) is the distance function and a, b, c, d are data objects. Such constraints are readily available in many problems where pairwise constraints are not natural to obtain. In this paper we consider the problem of learning a Mahalanobis distance metric from supervision in the form of relative distance comparisons. We propose a simple, yet effective, algorithm that minimizes a convex objective function corresponding to the sum of squared residuals of constraints. We also extend our model and algorithm to promote sparsity in the learned metric matrix. Experimental results suggest that our method consistently outperforms existing methods in terms of clustering accuracy. Furthermore, the sparsity extension leads to more stable estimation when the dimension is high and only a small amount of supervision is given.

Mixed-Criticality Scheduling upon Varying-Speed Processors

A varying-speed processor is characterized by two execution speeds: a normal speed and a degraded speed. Under normal circumstances it will execute at its normal speed, conditions during run-time may cause it to execute more slowly (but no slower than at its degraded speed). The problem of executing an integrated workload, consisting of some more important components and some less important ones, upon such a varying-speed processor is considered. It is desired that all components execute correctly under normal circumstances, whereas the more important components should execute correctly (although the less important components need not) if the processor runs at any speed no slower than its specified degraded speed.

Graph-regularized dual Lasso for robust eQTL mapping

Motivation: As a promising tool for dissecting the genetic basis of complex traits, expression quantitative trait loci (eQTL) mapping has attracted increasing research interest. An important issue in eQTL mapping is how to effectively integrate networks representing interactions among genetic markers and genes. Recently, several Lasso-based methods have been proposed to leverage such network information. Despite their success, existing methods have three common limitations: (i) a preprocessing step is usually needed to cluster the networks; (ii) the incompleteness of the networks and the noise in them are not considered; (iii) other available information, such as location of genetic markers and pathway information are not integrated. Results: To address the limitations of the existing methods, we propose Graph-regularized Dual Lasso (GDL), a robust approach for eQTL mapping. GDL integrates the correlation structures among genetic markers and traits simultaneously. It also takes into account the incompleteness of the networks and is robust to the noise. GDL utilizes graph-based regularizers to model the prior networks and does not require an explicit clustering step. Moreover, it enables further refinement of the partial and noisy networks. We further generalize GDL to incorporate the location of genetic makers and gene-pathway information. We perform extensive experimental evaluations using both simulated and real datasets. Experimental results demonstrate that the proposed methods can effectively integrate various available priori knowledge and significantly outperform the state-of-the-art eQTL mapping methods. Availability: Software for both C++ version and Matlab version is available at http://www.cs.unc.edu/∼weicheng/. Contact: weiwang@cs.ucla.edu Supplementary information: Supplementary data are available at Bioinformatics online.

MC-Fluid: Simplified and Optimally Quantified

The fluid scheduling model allows for schedules in which an individual task may be assigned a fraction of a processor at each time instant. These assignments are subject to the constraints that no fraction exceeds one and the sum of all the assigned fractions do not exceed the sum of the computing capacities of all the processors at any instant. An algorithm, MC-Fluid, has recently been proposed for scheduling systems of mixed-criticality implicit-deadline sporadic tasks under the fluid scheduling model. MC-Fluid has been shown to have a speedup bound no worse than (1 + √5)/2 or ≈ 1.618 for scheduling dual-criticality systems. We derive here a simplified variant of MC-Fluid called MCF, that has run-time linear in the number of tasks. We prove that this simplified variant has a speedup bound no worse than 4/3 for dual-criticality systems, and show that this implies that MC-Fluid, too, has a speedup bound no worse than 4/3. We know from prior results in uniprocessor mixed-criticality scheduling that no algorithm may have a speedup bound smaller than 4/3, allowing us to conclude that MCF and MC-Fluid are in fact speedup-optimal for dual-criticality scheduling.

Uniprocessor EDF scheduling of AVR task systems

The adaptive varying-rate (AVR) task model has been proposed as a means of modeling certain physically-derived constraints in CPSs in a manner that is more accurate (less pessimistic) than is possible using prior task models from real-time scheduling theory. Existing work on schedulability analysis of systems of AVR tasks is primarily restricted to fixed-priority scheduling; this paper establishes schedulability analysis results for systems of AVR and sporadic tasks under Earliest Deadline First (EDF) scheduling. The proposed analysis techniques are evaluated both theoretically via the speedup factor metric, and experimentally via schedulability experiments on randomly-generated task systems.