A5038278967- Robotic Mechanisms and Dynamics
- Neural Networks and Applications
- Robot Manipulation and Learning
- Neural dynamics and brain function
- Vehicle Routing Optimization Methods
- Advanced Algorithms and Applications
- Maritime Ports and Logistics
- Piezoelectric Actuators and Control
- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Adaptive Control of Nonlinear Systems
- Robotic Path Planning Algorithms
- Genomics and Rare Diseases
- Neuroscience and Neuropharmacology Research
- Image and Signal Denoising Methods
- Image and Video Stabilization
- Advanced Manufacturing and Logistics Optimization
- Gene Regulatory Network Analysis
- Image Processing Techniques and Applications
- Functional Brain Connectivity Studies
- Genomics and Phylogenetic Studies
- Advanced Measurement and Metrology Techniques
- Iterative Learning Control Systems
- Cell Image Analysis Techniques
- Genetic Associations and Epidemiology
Bellevue Hospital Center
2023
Allen Institute for Brain Science
2019-2020
Allen Institute
2019-2020
Inner Mongolia Electric Power (China)
2018-2019
University of Washington Medical Center
2017
University of Washington
2016-2017
University of Pittsburgh
2013-2016
University of Technology Sydney
2011-2013
Sun Yat-sen University
2008-2012
Academic Degrees & Graduate Education
2008
Different from gradient-based neural networks, a special kind of recurrent network (RNN) has recently been proposed by Zhang for online matrix inversion. Such an RNN is designed based on matrix-valued error function instead scalar-valued function. In addition, it was depicted in implicit dynamics explicit dynamics. this paper, we develop and investigate discrete-time model (termed as such abbreviated ZNN presentation convenience), which system difference equations. Comparing with Newton...
To solve the inverse kinematic problem of redundant robot manipulators, two redundancy-resolution schemes are investigated: one is resolved at joint-velocity level, and other joint-acceleration level. Both reformulated as a quadratic programming (QP) problem. Two recurrent neural networks (RNNs) then developed for online solution resultant QP The first RNN solver based on gradient-descent method termed gradient network (GNN). Zhang 's neural-dynamic (ZNN). computer simulations performed...
Experimental studies in neuroscience are producing data at a rapidly increasing rate, providing exciting opportunities and formidable challenges to existing theoretical modeling approaches. To turn massive datasets into predictive quantitative frameworks, the field needs software solutions for systematic integration of realistic, multiscale models. Here we describe Brain Modeling ToolKit (BMTK), suite building models performing simulations multiple levels resolution, from biophysically...
Short-chain acyl-coA dehydrogenase deficiency (SCADD) is an autosomal recessive inborn error of mitochondrial fatty acid oxidation caused by ACADS gene alterations. SCADD a heterogeneous condition, sometimes considered to be solely biochemical condition given that it has been associated with variable clinical phenotypes ranging from no symptoms or signs metabolic decompensation occurring early in life. A reason for this variability due SCAD alterations, such as the common p.Gly209Ser, confer...
Abstract In this paper, a bi‐criteria weighting scheme is proposed for the optimal motion control of redundant robot manipulators. To diminish discontinuity phenomenon pure infinity‐norm velocity minimization (INVM) scheme, redundancy‐resolution combines minimum kinetic energy and INVM via factor. Joint physical limits such as joint joint‐velocity could also be incorporated simultaneously into formulation. The kinematic can reformulated finally quadratic programming (QP) problem. As...
Abstract Background Ubiquitination is a very important process in protein post-translational modification, which has been widely investigated by biology scientists and researchers. Different experimental computational methods have developed to identify the ubiquitination sites sequences. This paper aims at exploring machine learning for prediction of using physicochemical properties (PCPs) amino acids Results We first establish six different data sets, whose records contain both...
In this paper, to diminish discontinuity points arising in the infinity-norm velocity minimization scheme, a bi-criteria scheme is presented based on new neural network solver (i.e., primal-dual linear variational inequalities (LVI)). Such kinematic control of redundant manipulators can incorporate joint physical limits such as and simultaneously. Moreover, be formulated quadratic programming (QP) problem. As real-time QP solver, LVI-based established with simple piecewise structure higher...
A multiobjective optimization model is presented in this paper for the Autonomous Straddle Carriers Scheduling (ASCS) problem automated container terminals, which more practical than single objective model. The considers three objectives [i.e., (SCs) traveling time, SC waiting time and finishing of high-priority container-transferring jobs], their weighted sum investigated as representative example. formulated a pickup delivery with windows form binary integer programming. An exact algorithm...
Although neural supersampling has achieved great success in various applications for improving image quality, it is still difficult to apply a wide range of real-time rendering due the high computational power demand. Most existing methods are computationally expensive and require high-performance hardware, preventing their use on platforms with limited such as smartphones. To this end, we propose new framework reconstruct high-quality out low-resolution one, which sufficiently lightweight...
A bstract Structural rules underlying functional properties of cortical circuits are poorly understood. To explore these systematically, we integrated information from extensive literature curation and large-scale experimental surveys into a data-driven, biologically realistic model the mouse primary visual cortex. The was constructed at two levels granularity, using either biophysically-detailed or point-neurons, with identical network connectivity. Both variants were compared to each other...
To obtain the inverses of time-varying matrices in real time, a special kind recurrent neural networks has recently been proposed by Zhang et al. It is proved that such network (ZNN) could globally exponentially converge to exact inverse given matrix. find out effect time-derivative term on global convergence as well for easier hardware-implementation purposes, ZNN model without exploiting information investigated this paper inverting online matrices. Theoretical results both constant matrix...
The advent of next-generation sequencing has dramatically decreased the cost for whole-genome and increased viability its application in research clinical care. Personal Genome Project (PGP) provides unrestricted access to genomes individuals their associated phenotypes. This resource enabled Critical Assessment Interpretation (CAGI) create a community challenge assess bioinformatics community's ability predict traits from whole genomes. In CAGI PGP challenge, researchers were asked whether...
Structural rules underlying functional properties of cortical circuits are poorly understood. To explore these systematically, we integrated information from extensive literature curation and large-scale experimental surveys into data-driven, biologically realistic models the mouse primary visual cortex. The were constructed at two levels granularity, using either biophysically-detailed or point-neuron models, with identical network connectivity. Both compared to each other recordings neural...
Objective The objective of this investigation is to evaluate binary prediction methods for predicting disease status using high-dimensional genomic data. central hypothesis that the Bayesian network (BN)-based method called efficient multivariate classifier (EBMC) will do well at task because EBMC builds on BN-based have performed learning epistatic interactions. Method We how eight perform discrete datasets containing are as follows: naive Bayes (NB), model averaging NB (MANB), feature...