Spatiotemporal Context Awareness for Urban Traffic Modeling and Prediction: Sparse Representation Based Variable Selection

Spatial-temporal correlations among the data play an important role in traffic flow prediction. Correspondingly, traffic modeling and prediction based on big data analytics emerges due to the city-scale interactions among traffic flows. A new methodology based on sparse representation is proposed to reveal the spatial-temporal dependencies among traffic flows so as to simplify the correlations among traffic data for the prediction task at a given sensor. Three important findings are observed in the experiments: (1) Only traffic flows immediately prior to the present time affect the formation of current traffic flows, which implies the possibility to reduce the traditional high-order predictors into an 1-order model. (2) The spatial context relevant to a given prediction task is more complex than what is assumed to exist locally and can spread out to the whole city. (3) The spatial context varies with the target sensor undergoing prediction and enlarges with the increment of time lag for prediction. Because the scope of human mobility is subject to travel time, identifying the varying spatial context against time lag is crucial for prediction. Since sparse representation can capture the varying spatial context to adapt to the prediction task, it outperforms the traditional methods the inputs of which are confined as the data from a fixed number of nearby sensors. As the spatial-temporal context for any prediction task is fully detected from the traffic data in an automated manner, where no additional information regarding network topology is needed, it has good scalability to be applicable to large-scale networks.     

Modeling DNA sequence-based cis-regulatory gene networks

Gene network analysis requires computationally based models which represent the functional architecture of regulatory interactions, and which provide directly testable predictions. The type of model that is useful is constrained by the particular features of developmentally active cis-regulatory systems. These systems function by processing diverse regulatory inputs, generating novel regulatory outputs. A computational model which explicitly accommodates this basic concept was developed earlier for the cis-regulatory system of the endo16 gene of the sea urchin. This model represents the genetically mandated logic functions that the system executes, but also shows how time-varying kinetic inputs are processed in different circumstances into particular kinetic outputs. The same basic design features can be utilized to construct models that connect the large number of cis-regulatory elements constituting developmental gene networks. The ultimate aim of the network models discussed here is to represent the regulatory relationships among the genomic control systems of the genes in the network, and to state their functional meaning. The target site sequences of the cis-regulatory elements of these genes constitute the physical basis of the network architecture. Useful models for developmental regulatory networks must represent the genetic logic by which the system operates, but must also be capable of explaining the real time dynamics of cis-regulatory response as kinetic input and output data become available. Most importantly, however, such models must display in a direct and transparent manner fundamental network design features such as intra- and intercellular feedback circuitry; the sources of parallel inputs into each cis-regulatory element; gene battery organization; and use of repressive spatial inputs in specification and boundary formation. Successful network models lead to direct tests of key architectural features by targeted cis-regulatory analysis.     

Dynamics and Energetics: A Consensus Analysis of the Impact of Calcium on EF-CaM Protein Complex

We have studied the relationship between dynamical correlations and energetic contributions in an attempt to model the transmission of information inside protein-protein complexes. The complex formed between the edema factor (EF) of Bacillus anthracis and calmodulin (CaM) was taken as an example, as the formation and stability of the complex depend on the calcium complexation level. The effect of calcium through EF-CaM residue network has been investigated with various approaches: 1), the elastic network model; 2), the local feature analysis; 3), the generalized correlations; and 4), the energetic dependency maps (EDMs), on 15-ns molecular dynamics simulations of the complex loaded with 0, 2, or 4 Ca²⁺ ions. The elastic network model correctly describes the basic architecture of the complex but is poorly sensitive to the level of calcium compared to the other methods. The local feature analysis allows us to characterize the local dynamics of the complex and the propagation of the calcium signal through CaM. The analyses of global dynamics and energetics—through generalized correlations and EDMs—provide a comprehensive picture of EF-CaM architecture and can be unified by using the concept of residue network connectedness. A medium connectedness, defined as the ability of each residue to communicate with all remaining parts of the complex, is observed for the 2Ca²⁺ level, which was experimentally identified as the most stable form of EF-CaM. The hierarchy of relative stabilities given by the EDMs sheds a new light on the EF-CaM interaction mechanism described experimentally and supports an organization of the complex architecture centered around nucleation points.     

Parallel and serial processes in the human oculomotor system: bimodal integration and express saccades

Saccadic reaction times (SRTs) were analyzed in the context of stochastic models of information processing (e.g., Townsend and Ashby 1983) to reveal the processing architecture(s) underlying integrative interactions between visual and auditory inputs and the mechanisms of express saccades. The results support the following conclusions. Bimodal (visual + auditory) targets are processed in parallel, and facilitate SRT to an extent that exceeds levels attainable by probability summation. This strongly implies neural summation between elements responding to spatially aligned visual and auditory inputs in the human oculomotor system. Second, express saccades are produced within a separable processing stage that is organized in series with that responsible for intersensory integration. A model is developed that implements this combination of parallel and serial processing. The activity in parallel input channels is summed within a sensory stage which is organized in series with a pre-motor and motor stage. The time course of each subprocess is considered a random variable, and different experimental manipulations can selectively influence different stages. Parallels between the model and physiological data are explored.     

Performance management of manufacturing system networks

Performance management of communication networks is critical for speed, reliability, and flexibility of information exchange between different components, subsystems, and sectors (e.g., factory, engineering design, and administration) of production process organizations in the environment of computer integrated manufacturing (CIM). Essential to this distributed total manufacturing system is the integrated communications network over which the information leading to process interactions and plant management and control is exchanged. Such a network must be capable of handling heterogeneous traffic resulting from intermachine communications at the factory floor, CAD drawings, design specifications, and administrative information. The objective is to improve the efficiency in handling various types of messages, e.g., control signals, sensor data, and production orders, by on-line adjustment of the parameters of the network protocol. This paper presents a conceptual design, development, and implementation of a network performance management scheme for CIM applications including flexible manufacturing. The performance management algorithm is formulated using the concepts of: (1) Perturbation analysis of discrete event dynamic systems; (2) stochastic approximation; and (3) learning automata. The proposed concept for performance management can also serve as a general framework to assist design, operation, and management of flexible manufacturing systems. The performance management procedure has been tested via emulation on a network test bed that is based on the manufacturing automation protocol (MAP) which has been widely used for CIM networking. The conceptual design presented in this paper offers a step forward to bridging the gap between management standards and users' demands for efficient network operations since most standards such as ISO and IEEE address only the architecture, services, and interfaces for network management.     

A model for integrating elementary neural functions into delayed-response behavior

It is well established that various cortical regions can implement a wide array of neural processes, yet the mechanisms which integrate these processes into behavior-producing, brain-scale activity remain elusive. We propose that an important role in this respect might be played by executive structures controlling the traffic of information between the cortical regions involved. To illustrate this hypothesis, we present a neural network model comprising a set of interconnected structures harboring stimulus-related activity (visual representation, working memory, and planning), and a group of executive units with task-related activity patterns that manage the information flowing between them. The resulting dynamics allows the network to perform the dual task of either retaining an image during a delay (delayed-matching to sample task), or recalling from this image another one that has been associated with it during training (delayed-pair association task). The model reproduces behavioral and electrophysiological data gathered on the inferior temporal and prefrontal cortices of primates performing these same tasks. It also makes predictions on how neural activity coding for the recall of the image associated with the sample emerges and becomes prospective during the training phase. The network dynamics proves to be very stable against perturbations, and it exhibits signs of scale-invariant organization and cooperativity. The present network represents a possible neural implementation for active, top-down, prospective memory retrieval in primates. The model suggests that brain activity leading to performance of cognitive tasks might be organized in modular fashion, simple neural functions becoming integrated into more complex behavior by executive structures harbored in prefrontal cortex and/or basal ganglia.     

A continuous-time neural model for sequential action

Action selection, planning and execution are continuous processes that evolve over time, responding to perceptual feedback as well as evolving top-down constraints. Existing models of routine sequential action (e.g. coffee- or pancake-making) generally fall into one of two classes: hierarchical models that include hand-built task representations, or heterarchical models that must learn to represent hierarchy via temporal context, but thus far lack goal-orientedness. We present a biologically motivated model of the latter class that, because it is situated in the Leabra neural architecture, affords an opportunity to include both unsupervised and goal-directed learning mechanisms. Moreover, we embed this neurocomputational model in the theoretical framework of the theory of event coding (TEC), which posits that actions and perceptions share a common representation with bidirectional associations between the two. Thus, in this view, not only does perception select actions (along with task context), but actions are also used to generate perceptions (i.e. intended effects). We propose a neural model that implements TEC to carry out sequential action control in hierarchically structured tasks such as coffee-making. Unlike traditional feedforward discrete-time neural network models, which use static percepts to generate static outputs, our biological model accepts continuous-time inputs and likewise generates non-stationary outputs, making short-timescale dynamic predictions.     

A Cognitive Neural Architecture Able to Learn and Communicate through Natural Language

Communicative interactions involve a kind of procedural knowledge that is used by the human brain for processing verbal and nonverbal inputs and for language production. Although considerable work has been done on modeling human language abilities, it has been difficult to bring them together to a comprehensive tabula rasa system compatible with current knowledge of how verbal information is processed in the brain. This work presents a cognitive system, entirely based on a large-scale neural architecture, which was developed to shed light on the procedural knowledge involved in language elaboration. The main component of this system is the central executive, which is a supervising system that coordinates the other components of the working memory. In our model, the central executive is a neural network that takes as input the neural activation states of the short-term memory and yields as output mental actions, which control the flow of information among the working memory components through neural gating mechanisms. The proposed system is capable of learning to communicate through natural language starting from tabula rasa, without any a priori knowledge of the structure of phrases, meaning of words, role of the different classes of words, only by interacting with a human through a text-based interface, using an open-ended incremental learning process. It is able to learn nouns, verbs, adjectives, pronouns and other word classes, and to use them in expressive language. The model was validated on a corpus of 1587 input sentences, based on literature on early language assessment, at the level of about 4-years old child, and produced 521 output sentences, expressing a broad range of language processing functionalities.     

Leveraging Non-Targeted Metabolite Profiling via Statistical Genomics

One of the challenges of systems biology is to integrate multiple sources of data in order to build a cohesive view of the system of study. Here we describe the mass spectrometry based profiling of maize kernels, a model system for genomic studies and a cornerstone of the agroeconomy. Using a network analysis, we can include 97.5% of the 8,710 features detected from 210 varieties into a single framework. More conservatively, 47.1% of compounds detected can be organized into a network with 48 distinct modules. Eigenvalues were calculated for each module and then used as inputs for genome-wide association studies. Nineteen modules returned significant results, illustrating the genetic control of biochemical networks within the maize kernel. Our approach leverages the correlations between the genome and metabolome to mutually enhance their annotation and thus enable biological interpretation. This method is applicable to any organism with sufficient bioinformatic resources.     

Towards organizing health knowledge on community-based health services

Online community-based health services accumulate a huge amount of unstructured health question answering (QA) records at a continuously increasing pace. The ability to organize these health QA records has been found to be effective for data access. The existing approaches for organizing information are often not applicable to health domain due to its domain nature as characterized by complex relation among entities, large vocabulary gap, and heterogeneity of users. To tackle these challenges, we propose a top-down organization scheme, which can automatically assign the unstructured health-related records into a hierarchy with prior domain knowledge. Besides automatic hierarchy prototype generation, it also enables each data instance to be associated with multiple leaf nodes and profiles each node with terminologies. Based on this scheme, we design a hierarchy-based health information retrieval system. Experiments on a real-world dataset demonstrate the effectiveness of our scheme in organizing health QA into a topic hierarchy and retrieving health QA records from the topic hierarchy.     

Computer modeling of mechanisms of the information processing in the olfactory bulb. I. Model of the structure-functional organizations of neuronal elements in the olfactory bulb and receptor epithelium

A computer model of the olfactory bulb was constructed. The paper describes: 1) the general architecture of a model neuron network that reflects the neurophysiological experimental and theoretical data on the structural and functional organization of the peripheral part of the olfactory system, the olfactory bulb with inputs from olfactory receptor neurons; 2) the organization of each of three levels of the model: receptors, olfactory glomeruli, and basic neurons; and 3) a scenario of the computer model work. In some aspects, in particular, in the principle of information presentation, the treatment of the role of basic neurons (mitral and tufted cells), and their interrelations in modules, the model favorably differs from the available olfactory bulb models. The model is basic and provides further refinement of the architecture, an increase in the number of modules, and the modeling of the learning process.     

Thinking about Bacillus subtilis as a multicellular organism

Initial attempts to use colony morphogenesis as a tool to investigate bacterial multicellularity were limited by the fact that laboratory strains often have lost many of their developmental properties. Recent advances in elucidating the molecular mechanisms underlying colony morphogenesis have been made possible through the use of undomesticated strains. In particular, Bacillus subtilis has proven to be a remarkable model system to study colony morphogenesis because of its well-characterized developmental features. Genetic screens that analyze mutants defective in colony morphology have led to the discovery of an intricate regulatory network that controls the production of an extracellular matrix. This matrix is essential for the development of complex colony architecture characterized by aerial projections that serve as preferential sites for sporulation. While much progress has been made, the challenge for future studies will be to determine the underlying mechanisms that regulate development such that differentiation occurs in a spatially and temporally organized manner.     

Molecular architecture and assembly of the yeast kinetochore MIND complex

The MIND multiprotein complex is a conserved, essential component of eukaryotic kinetochores and is a constituent of the tripartite KMN network that directly attaches the kinetochore to the mitotic spindle. The primary microtubule-binding complex in this network, NDC80, has been extensively characterized, but very little is known about the structure or function of the MIND complex. In this study, we present biochemical, hydrodynamic, electron microscopy, and small-angle x-ray scattering data that provide insight into the overall architecture and assembly of the MIND complex and the physical relationship of the complex with other components of the KMN network. We propose a model for the overall structure of the complex and provide data on the interactions with NDC80, Spc105p, and thus the mitotic spindle.     

Using historical data for bioprocess optimization: modeling wine characteristics using artificial neural networks and archived process information

Optimization of fermentation processes is a difficult task that relies on an understanding of the complex effects of processing inputs on productivity and quality outputs. Because of the complexity of these biological systems, traditional optimization methods utilizing mathematical models and statistically designed experiments are less effective, especially on a production scale. At the same time, information is being collected on a regular basis during the course of normal manufacturing and process development that is rarely fully utilized. We are developing an optimization method in which historical process data is used to train an artificial neural network for correlation of processing inputs and outputs. Subsequently, an optimization routine is used in conjunction with the trained neural network to find optimal processing conditions given the desired product characteristics and any constraints on inputs. Wine processing is being used as a case study for this work. Using data from wine produced in our pilot winery over the past 3 years, we have demonstrated that trained neural networks can be used successfully to predict the yeast-fermentation kinetics, as well as chemical and sensory properties of the finished wine, based solely on the properties of the grapes and the intended processing. To accomplish this, a hybrid neural network training method, Stop Training with Validation (STV), has been developed to find the most desirable neural network architecture and training level. As industrial historical data will not be evenly spaced over the entire possible search space, we have also investigated the ability of the trained neural networks to interpolate and extrapolate with data not used during training. Because a company will utilize its own existing process data for this method, the result of this work will be a general fermentation optimization method that can be applied to fermentation processes to improve quality and productivity.     

Analysis of the Neural Network Regulating the Cardio-Renal System in the Central Nervous System of Helix pomatia L.

In the central nervous system of Helix pomatia 22 cells takingpart in the regulation of the cardio-renal system have beenidentified. These cells are scattered throughout the visceraland right parietal ganglia. Among identified cells sensory,motor, and interneurons were found and the hierarchical characterof the network was stated. The network regulating heart activitywas found to be of the over-guaranteed, convergent type, wherethe inputs predominate and the outputs form independent, parallelpathways, being coordinated by interneurons. This neural networkcan be divided into two levels: collectors and coordinators,which are responsible for the conduction or analysis of theafferent inputs, respectively. The localization of identifiedcells was studied by intracellular and retrograde injectionof CoCl₂. Primary sensory cells were found to be bipolar, motoneuronesand interneurones were unipolar or pseudo-unipolar with richarborization within the ganglia. Interneuron V21 showed a phasic or tonic pattern of firing.The phasic activity of the neuron V21 appeared as a burst correlatedwith the individual heartbeats. The tonic pattern of firingwas caused by activation of various inputs of the cardio-renalsystem and led always to the stopping of the heartbeat. Thenetwork regulating the heartbeat was organized around the interneurones.In the activity pattern of the interneurones the origin of theinput was not distinguishable; however, on the middle levelof the system it can be verified. These cells play a role inthe storage of information originating Irom various receptorareas. According to this the regulatory network can be dividedinto subsystems.     

Reinforcement Learning on Slow Features of High-Dimensional Input Streams

Humans and animals are able to learn complex behaviors based on a massive stream of sensory information from different modalities. Early animal studies have identified learning mechanisms that are based on reward and punishment such that animals tend to avoid actions that lead to punishment whereas rewarded actions are reinforced. However, most algorithms for reward-based learning are only applicable if the dimensionality of the state-space is sufficiently small or its structure is sufficiently simple. Therefore, the question arises how the problem of learning on high-dimensional data is solved in the brain. In this article, we propose a biologically plausible generic two-stage learning system that can directly be applied to raw high-dimensional input streams. The system is composed of a hierarchical slow feature analysis (SFA) network for preprocessing and a simple neural network on top that is trained based on rewards. We demonstrate by computer simulations that this generic architecture is able to learn quite demanding reinforcement learning tasks on high-dimensional visual input streams in a time that is comparable to the time needed when an explicit highly informative low-dimensional state-space representation is given instead of the high-dimensional visual input. The learning speed of the proposed architecture in a task similar to the Morris water maze task is comparable to that found in experimental studies with rats. This study thus supports the hypothesis that slowness learning is one important unsupervised learning principle utilized in the brain to form efficient state representations for behavioral learning.     

A network capable of reading the temporal codes in the olfactory discrimination of insects]

Recent work on the insect olfactory system has shown that its mushroom bodies (one of its major components) are involved in the fine discrimination of odours and that the temporal organisation of spike discharges plays a fundamental role. We propose here a model of a network that is able to decode the temporal patterns which characterise an odour. This model has three fundamental properties that seem to exist in all mushroom bodies of insects studied so far: a) long lasting inhibitions with rebounds, able to facilitate delayed spike generation; b) synaptic plasticity, which allows the network to learn to recognise temporal patterns; c) above all a large interconnection, which allows this network to recognise intervals of various duration. This model thus appears suited to identify combinations of temporal patterns in the dendrites of Kenyon cells (the principal cells in the calyces of the mushroom bodies). Moreover, the mushroom bodies integrate multimodal inputs, suggesting that the detection of temporal patterns may be extended to the detection of a complex environment, combining in particular olfactive and visual inputs.     

Characterization of arabinoxylan/cellulose nanocrystals gels to investigate fluorescent probes mobility in bioinspired models of plant secondary cell wall

Biomass from lignocellulose (LC) is a highly complex network of cellulose, hemicellulose, and lignin, which is considered to be a sustainable source of fuels, chemicals and materials. To achieve an environmental friendly and efficient LC upgrading, a better understanding of the LC architecture is necessary. We have devised some LC bioinspired model systems, based on arabinoxylan gels, in which mobility of dextrans and BSA grafted with FITC has been studied by FRAP. Our results indicate that the probes diffusion is more influenced by their hydrodynamic radius than by the gel mesh size. The addition of some cellulose nanocrystals (CNCs) decreases polymer chain mobility and has low effect on the probes diffusion, suggesting that the gels are better organized in the presence of CNCs, as shown by rheological measurements and scanning electronic microscopy observations. This demonstrates that the FRAP analysis can be a powerful tool to screen the architecture of LC model systems.