DOCSLIB.ORG

Spatiotemporal Pattern Queries

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text

Load more

Spatiotemporal Pattern Queries Von der Fakult¨atf¨urMathematik und Informatik der FernUniversit¨atin Hagen genehmigte Dissertation zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.) von Mahmoud Attia Sakr Geburtsort: El-Scharkia, Agypten¨ Hagen, 2012 ii Dedicated to the martyrs of the Egyptian revolution Jan 25th 2011. To those who gave their lives to make Egypt a better place for everyone. i ii Aknowledgement goal goal start start When people see my work, they think I have reached these results like this. But actually it was like this! I guess this drawing holds for all research projects. When I saw it some years ago on Facebook, I thought this is really expressive! It was fun going the whole way to finishing this thesis. Without the support of my work group, family and friends, it would not have been possible to reach this end. I gratefully thank all of them and wish them all the best! All thanks and praise to God who has planned all this for me. My adviser Prof. Ralf Hartmut G¨utinghas been actively involved at all stages of this work from writing the proposal to finishing the thesis. He was always available for discussing research issues. I admire his ability to establish such a good relationship with foreign students. Very shortly after I joined his group, he introduced me to his family and friends. Finally, he is one of the few people I know who could beat me at table tennis! Thank you, Hartmut, for all your support, for the friendship, and for being a great teacher! I thank my colleagues in the work group for their patience during the long hours they spent with me helping with my research. I surely needed all the goodwill they have shown towards me. Thanks to Prof. Michael Gertz who accepted to review this thesis, and did the review in a short time. This was essential to me, as I was committed to a limited stay in Germany. I also need to express all my love to my family and to my son. They have carried on my responsibilities in Egypt during my stay in Germany. Despite the distance, they managed to always let me know they were continuously supporting me. iii iv AKNOWLEDGEMENT This work was funded in the first two years by DAAD, with supplemen- tary funds from the Egyptian study mission. The remaining 21 months were funded by the FernUniversit¨atin Hagen, Lehrgebiet Datenbanksysteme f¨ur neue Anwendungen, Prof. Ralf Hartmut G¨uting. Contents i Aknowledgement iii List of Tables ix List of Algorithms xi List of Figures xiii Abstract xv 1 Introduction 1 1.1 Problem Definition . .2 1.2 Thesis Contributions . .3 1.2.1 Individual STP Queries . .4 1.2.2 Group STP Queries . .6 1.2.3 Contributions . 10 1.3 Thesis Organization . 11 2 Related Work 13 2.1 Individual Spatiotemporal Pattern Queries . 13 2.2 Group Spatiotemporal Pattern Queries . 15 3 The Moving Objects Data Model (Review) 19 3.1 The Type System . 19 3.2 SECONDO . 22 3.3 The Query Language . 22 3.4 Time-Dependent Predicates . 24 4 Individual Spatiotemporal Patterns 27 4.1 The Spatiotemporal Pattern Predicate . 29 4.2 Evaluating The STP Predicate . 32 4.3 Extending The Definition Of The STP Predicate . 36 4.4 Optimizing The STP Predicate . 38 4.4.1 Basic Assumptions . 38 4.4.2 Query Optimization . 39 4.5 The Implementation In Secondo ................ 42 4.5.1 Extending The Kernel . 42 v vi CONTENTS 4.5.2 Extending The Optimizer . 43 4.6 Experimental Evaluation . 44 4.6.1 Experiment 1: The Overhead Of Evaluating The STP Predicate . 45 4.6.2 Experiment 2: Run-Time And Optimization Gain . 47 4.6.3 Experiment 3: Scalability . 49 4.7 Conclusions . 53 5 Group Spatiotemporal Patterns 55 5.1 Extending The Type System . 55 5.2 Group Spatiotemporal Pattern Operators . 56 5.2.1 The reportpattern Operator . 59 5.2.2 The gpattern Operator . 60 5.2.3 The crosspattern Operator . 63 5.3 The crosspattern Evaluation Algorithm . 65 5.3.1 Related Work (REVIEW) . 66 5.3.2 The Proposed Algorithm . 67 5.4 Examples . 76 5.5 Optimization . 78 5.5.1 Optimizing The gpattern Operator . 79 5.5.2 Optimizing The crosspattern Operator . 81 5.5.3 Optimizing The reportpattern Operator . 81 5.5.4 Notes On The Implementation . 82 5.6 Experimental Evaluation . 82 5.6.1 Evaluating The gpattern Operator . 82 5.6.2 Evaluating The crosspattern Operator . 86 5.6.3 A Remark . 90 5.7 Conclusions . 91 6 Temporal Reasoning 93 6.1 Temporal Reasoning (REVIEW) . 94 6.2 The Reasoning Task . 95 6.3 Integration With The Individual STP Operators . 97 6.4 Experimental Evaluation . 102 6.4.1 Experiment 1: The Overhead Of Temporal Reasoning . 102 6.4.2 Experiment 2: The Performance Gain Of Temporal Reasoning . 102 6.5 Discussion And Conclusions . 107 7 Application Examples 109 7.1 Air Traffic Control . 109 7.1.1 Missed Approach . 111 7.1.2 Stepwise-Descent Landing . 113 7.2 Ghent City Game . 115 8 Conclusions and Future Work 121 Appendices 123 CONTENTS vii A Experimental Repeatability 125 A.1 Repeating The Individual STP Experiments . 125 A.1.1 Repeating the First Experiment . 125 A.1.2 Repeating the Second Experiment . 126 viii CONTENTS List of Tables 2.1 Examples of the Group STPs . 16 3.1 The type system . 20 4.1 Individual spatiotemporal pattern operators . 28 4.2 A language for expressing interval relationships . 30 4.3 Mapping time-dependent predicates into standard predicates. 41 4.4 The database relations used in the scalability experiment . 50 5.1 The extended type system . 56 5.2 Group spatiotemporal pattern operators . 58 5.3 Composing component states . 72 5.4 Component states occurrence frequencies . 72 5.5 The datasets used for evaluating the gpattern operator . 83 5.6 Description of the datasets of the crosspattern experiment . 89 6.1 Individual STP operators with temporal reasoning . 101 A.1 The schema of the STPQExpr1Result relation . 126 A.2 The schemas of the Expr2StatsDO.csv and Expr2StatsEO.csv files . 127 ix x LIST OF TABLES List of Algorithms 1 The MatchPattern Algorithm . 34 2 The Extend Helper Function . 35 3 The gpattern Algorithm . 63 4 The mbool2mset Helper Function . 63 5 The DeleteUnitsBelowSize Helper Function . 63 6 The DeleteElemsBelowDuration Helper Function . 64 7 The crosspattern Algorithm . 68 8 The FindLargeConnectedComponents Helper Function . 69 9 The FindConnectedComponents Helper Function . 70 10 The IA-To-PA Helper Function . 96 11 MatchPattern With Temporal Reasoning . 98 12 The GetConsistentPeriods Helper Function . 99 xi xii LIST OF ALGORITHMS List of Figures 3.1 The sliced representation of a moving point . 22 3.2 Time-dependent predicates . 25 4.1 Evaluating the STP predicate . 32 4.2 The overhead of evaluating the STP predicate . 47 4.3 The run-time of the STP queries on the Trains20 relation . 49 4.4 Scalability results . 51 5.1 State NewlyAdded followed by state RemoveNow ....... 72 5.2 State NewlyAdded followed by state Split ............ 72 5.3 State Merged followed by state Split .............. 73 5.4 Helper indexes for representing the nodes of the pattern graph 75 5.5 The run-time of the gpattern operator . 85 5.6 The run-time of the crosspattern operator . 90 6.1 Continuous and discontinuous endpoint relationships . 95 6.2 The Decompose vector . 96 6.3 The j operator . 97 6.4 The overhead of temporal reasoning . 103 6.5 The performance gain of temporal reasoning . 105 6.6 Run-time statistics . 106 7.1 Integrating SECONDO with V-Analytics . 111 7.2 Missed approach query results . 113 7.3 Stepwise-descent query results . 115 xiii xiv LIST OF FIGURES Abstract Capturing moving objects data is now possible and becoming cheaper with the advances in positioning and sensor technologies. The increasing amount of such data and the various fields of applications call for intensive research work for building spatiotemporal database management systems. This in- volves several tasks such as modeling the moving objects data, designing query methods, query optimization, etc. This PhD project goes in this direc- tion. Our goal is to design and implement query operators for the Spatiotem- poral Pattern Queries (STP queries). This includes the design of expressive query operators, the integration with the query optimizers for efficient evalu- ation, and the implementation within the context of a spatiotemporal DBMS. In this thesis we distinguish between two kinds of spatiotemporal pat- terns: individual and group, according to the number of objects composing the pattern. Accordingly, two sets of query operators are proposed. An individual STP is a certain movement profile shown by the moving object during its observation period. An example of this is the pattern of a drifting ship. When a ship has engine problems, and the crew has no more control over it, it starts drifting. It moves according to the wind and current. This movement pattern is characterized by low speed and high drifting angle (the difference between the heading and the course over ground). Another example is the holding pattern of aircraft. When the run-way is busy and the landing of an aircraft has to be delayed, it is directed to join the holding pattern. The aircraft flies in circles over a pre-defined area in the airspace of the destination airport until the run-way is ready. The first part of this work proposes query operators for expressing such patterns and matching them on trajectory databases. They allow one to specify temporal order constraints on the fulfillment of several predicates.
Recommended publications
  • Odd-Paired Controls Frequency Doubling in Drosophila Segmentation by Altering the Pair-Rule Gene Regulatory Network Erik Clark*, Michael Akam

    Odd-Paired Controls Frequency Doubling in Drosophila Segmentation by Altering the Pair-Rule Gene Regulatory Network Erik Clark*, Michael Akam

    RESEARCH ARTICLE Odd-paired controls frequency doubling in Drosophila segmentation by altering the pair-rule gene regulatory network Erik Clark*, Michael Akam Laboratory for Development and Evolution, Department of Zoology, University of Cambridge, Cambridge, United Kingdom Abstract The Drosophila embryo transiently exhibits a double-segment periodicity, defined by the expression of seven ’pair-rule’ genes, each in a pattern of seven stripes. At gastrulation, interactions between the pair-rule genes lead to frequency doubling and the patterning of 14 parasegment boundaries. In contrast to earlier stages of Drosophila anteroposterior patterning, this transition is not well understood. By carefully analysing the spatiotemporal dynamics of pair- rule gene expression, we demonstrate that frequency-doubling is precipitated by multiple coordinated changes to the network of regulatory interactions between the pair-rule genes. We identify the broadly expressed but temporally patterned transcription factor, Odd-paired (Opa/ Zic), as the cause of these changes, and show that the patterning of the even-numbered parasegment boundaries relies on Opa-dependent regulatory interactions. Our findings indicate that the pair-rule gene regulatory network has a temporally modulated topology, permitting the pair-rule genes to play stage-specific patterning roles. DOI: 10.7554/eLife.18215.001 Introduction *For correspondence: ec491@ Segmentation is a developmental process that subdivides an animal body axis into similar, repeating cam.ac.uk units (Hannibal and Patel, 2013). Segmentation of the main body axis underlies the body plans of Competing interests: The arthropods, annelids and vertebrates (Telford et al., 2008; Balavoine, 2014; Graham et al., 2014). authors declare that no In arthropods, segmentation first involves setting up polarised boundaries early in development to competing interests exist.
  • Spatiotemporal Dynamics of Continuum Neural Fields

    Spatiotemporal Dynamics of Continuum Neural Fields

    Home Search Collections Journals About Contact us My IOPscience Spatiotemporal dynamics of continuum neural fields This article has been downloaded from IOPscience. Please scroll down to see the full text article. 2012 J. Phys. A: Math. Theor. 45 033001 (http://iopscience.iop.org/1751-8121/45/3/033001) View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: 67.2.165.135 The article was downloaded on 14/12/2011 at 20:02 Please note that terms and conditions apply. IOP PUBLISHING JOURNAL OF PHYSICS A: MATHEMATICAL AND THEORETICAL J. Phys. A: Math. Theor. 45 (2012) 033001 (109pp) doi:10.1088/1751-8113/45/3/033001 TOPICAL REVIEW Spatiotemporal dynamics of continuum neural fields Paul C Bressloff Department of Mathematics, University of Utah, 155 South 1400 East, Salt Lake City, UT 84112, USA E-mail: [email protected] Received 21 July 2011, in final form 11 November 2011 Published 14 December 2011 Online at stacks.iop.org/JPhysA/45/033001 Abstract We survey recent analytical approaches to studying the spatiotemporal dynamics of continuum neural fields. Neural fields model the large-scale dynamics of spatially structured biological neural networks in terms of nonlinear integrodifferential equations whose associated integral kernels represent the spatial distribution of neuronal synaptic connections. They provide an important example of spatially extended excitable systems with nonlocal interactions and exhibit a wide range of spatially coherent dynamics including traveling waves oscillations and Turing-like patterns. PACS numbers: 87.19.L−, 87.19.lj, 87.19.lp, 87.19.lq, 87.10.Ed, 05.40.−a (Some figures may appear in colour only in the online journal) Contents 1.
  • Spatiotemporal Neural Pattern Similarity Supports Episodic Memory

    Spatiotemporal Neural Pattern Similarity Supports Episodic Memory

    Report Spatiotemporal Neural Pattern Similarity Supports Episodic Memory Highlights Authors d EEG STPS during encoding predicted later memory Yi Lu, Changming Wang, Chuansheng Chen, Gui Xue d Item-specific STPS occurred around 500 ms after stimulus onset Correspondence [email protected] d LPFC anodal tDCS enhanced subsequent memory and item- specific STPS In Brief Using EEG and tDCS, Lu et al. examined the neural mechanisms of memory formation. They found that greater item- specific STPS during encoding predicted better later memory. Anodal tDCS over the left LPFC specifically enhanced STPS and memory. Lu et al., 2015, Current Biology 25, 780–785 March 16, 2015 ª2015 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.cub.2015.01.055 Current Biology 25, 780–785, March 16, 2015 ª2015 Elsevier Ltd All rights reserved http://dx.doi.org/10.1016/j.cub.2015.01.055 Report Spatiotemporal Neural Pattern Similarity Supports Episodic Memory Yi Lu,1,2 Changming Wang,1,2 Chuansheng Chen,3 We recorded EEG while 20 participants were studying the vi- and Gui Xue1,2,* sual forms of 120 novel visual symbols (i.e., Korean Hangul char- 1State Key Laboratory of Cognitive Neuroscience and acters), using a visual structure judgment task (Figure 1A). Each Learning and IDG/McGovern Institute for Brain Research, character was repeated three times, with an inter-repetition-in- Beijing Normal University, Beijing 100875, China terval (IRI) of 4–7 trials. Their recognition memory was probed 2Center for Collaboration and Innovation in Brain and Learning 1 day later using a six-point old or new judgment task (Fig- Sciences, Beijing Normal University, Beijing 100875, China ure 1B).
  • Odd-Paired Controls Frequency Doubling in Drosophila Segmentation by Altering the Pair-Rule Gene Regulatory Network

    Odd-Paired Controls Frequency Doubling in Drosophila Segmentation by Altering the Pair-Rule Gene Regulatory Network

    Odd-paired controls frequency doubling in Drosophila segmentation by altering the pair-rule gene regulatory network Erik Clark* and Michael Akam Laboratory for Development and Evolution, Department of Zoology, University of Cambridge, Downing Street, Cambridge CB2 3EJ, United Kingdom *Correspondence to: [email protected] The authors declare that no competing interests exist. 1 ABSTRACT 2 3 The Drosophila embryo transiently exhibits a double segment periodicity, defined by the expression of seven 4 “pair-rule” genes, each in a pattern of seven stripes. At gastrulation, interactions between the pair-rule genes 5 lead to frequency doubling and the patterning of fourteen parasegment boundaries. In contrast to earlier 6 stages of Drosophila anteroposterior patterning, this transition is not well understood. By carefully analysing 7 the spatiotemporal dynamics of pair-rule gene expression, we demonstrate that frequency-doubling is 8 precipitated by multiple coordinated changes to the network of regulatory interactions between the pair-rule 9 genes. We identify the broadly expressed but temporally patterned transcription factor, Odd-paired 10 (Opa/Zic), as the cause of these changes, and show that the patterning of the even-numbered parasegment 11 boundaries relies on Opa-dependent regulatory interactions. Our findings indicate that the pair-rule gene 12 regulatory network has a temporally-modulated topology, permitting the pair-rule genes to play stage- 13 specific patterning roles. 14 15 Keywords: pair-rule genes; segmentation; Drosophila; patterning; gene regulatory network; Odd-paired; Zic 16 17 18 INTRODUCTION 19 20 Segmentation is a developmental process that subdivides an animal body axis into similar, repeating units 21 (Hannibal & Patel 2013).
  • Learning, Recognizing and Early Classification of Spatio-Temporal Patterns Using Spike Timing Neural Networks

    Learning, Recognizing and Early Classification of Spatio-Temporal Patterns Using Spike Timing Neural Networks

    University of Nevada, Reno Learning, Recognizing and Early Classification of Spatio-Temporal Patterns using Spike Timing Neural Networks A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Computer Science and Engineering by Banafsheh Rekabdar Dr. Monica Nicolescu/Thesis Advisor Dr. Mircea Nicolescu/Thesis Co-Advisor May 2017 THE GRADUATE SCHOOL We recommend that the dissertation prepared under our supervision by BANAFSHEH REKABDAR Entitled Learning, Recognizing and Early Classification Of Spatio-Temporal Patterns Using Spike Timing Neural Networks be accepted in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Monica Nicolescu, Ph.D., Advisor Mircea Nicolescu, Ph.D., Committee Member George Bebis, Ph.D., Committee Member Sushil Louis, Ph.D., Committee Member Raul Rojas, Ph.D., Graduate School Representative David W. Zeh, Ph. D., Dean, Graduate School May, 2017 i Abstract by Banafsheh Rekabdar Learning and recognizing spatio-temporal patterns is an important problem for all biological systems. Gestures, movements, activities, all encompass both spatial and temporal information that is critical for implicit communication and learning. This dissertation presents a novel, unsupervised approach for learning, recognizing and early classifying spatio-temporal patterns using spiking neural networks for human- robotic domains. The proposed spiking approach has five variations which have been validated on images of handwritten digits and human hand gestures and motions. The main contributions of this work are as follows: i) it requires a very small number of training examples, ii) it enables early recognition from only partial information of the pattern, iii) it learns patterns in an unsupervised manner, iv) it accepts variable sized input patterns, v) it is invariant to scale and translation, vi) it can recognize pat- terns in real-time and, vii) it is suitable for human-robot interaction applications and has been successfully tested on a PR2 robot.
  • Controlling Spatiotemporal Pattern Formation in a Concentration Gradient with a Synthetic Toggle Switch

    Controlling Spatiotemporal Pattern Formation in a Concentration Gradient with a Synthetic Toggle Switch

    bioRxiv preprint doi: https://doi.org/10.1101/849711; this version posted November 21, 2019. The copyright holder for this preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under aCC-BY-NC 4.0 International license. Controlling spatiotemporal pattern formation in a concentration gradient with a synthetic toggle switch Içvara Barbier1, Rubén Perez Carrasco2* and Yolanda Schaerli1* 1 Department of Fundamental Microbiology, University of Lausanne, Biophore Building, 1015 Lausanne, Switzerland 2 Department of Mathematics, University College London, Gower Street, WC1E 6BT London, United Kingdom *Correspondence: [email protected] and [email protected] Abstract The formation of spatiotemporal patterns of gene expression is frequently guided by gradients of diffusible signaling molecules. The toggle switch subnetwork, composed of two cross-repressing transcription factors, is a common component of gene regulatory networks in charge of patterning, converting the continuous information provided by the gradient into discrete abutting stripes of gene expression. We present a synthetic biology framework to understand and characterize the spatiotemporal patterning properties of the toggle switch. To this end, we built a synthetic toggle switch controllable by diffusible molecules in Escherichia coli. We analyzed the patterning capabilities of the circuit by combining quantitative measurements with a mathematical reconstruction of the underlying dynamical system. The toggle switch can produce robust patterns with sharp boundaries, governed by bistability and hysteresis. We further demonstrate how the hysteresis, position, timing, and precision of the boundary can be controlled, highlighting the dynamical flexibility of the circuit.
  • Controlling Spatiotemporal Pattern Formation in a Concentration Gradient with a Synthetic Toggle Switch

    Controlling Spatiotemporal Pattern Formation in a Concentration Gradient with a Synthetic Toggle Switch

    bioRxiv preprint first posted online Nov. 21, 2019; doi: http://dx.doi.org/10.1101/849711. The copyright holder for this preprint (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC 4.0 International license. Controlling spatiotemporal pattern formation in a concentration gradient with a synthetic toggle switch Içvara Barbier1, Rubén Perez Carrasco2* and Yolanda Schaerli1* 1 Department of Fundamental Microbiology, University of Lausanne, Biophore Building, 1015 Lausanne, Switzerland 2 Department of Mathematics, University College London, Gower Street, WC1E 6BT London, United Kingdom *Correspondence: [email protected] and [email protected] Abstract The formation of spatiotemporal patterns of gene expression is frequently guided by gradients of diffusible signaling molecules. The toggle switch subnetwork, composed of two cross-repressing transcription factors, is a common component of gene regulatory networks in charge of patterning, converting the continuous information provided by the gradient into discrete abutting stripes of gene expression. We present a synthetic biology framework to understand and characterize the spatiotemporal patterning properties of the toggle switch. To this end, we built a synthetic toggle switch controllable by diffusible molecules in Escherichia coli. We analyzed the patterning capabilities of the circuit by combining quantitative measurements with a mathematical reconstruction of the underlying dynamical system. The toggle switch can produce robust patterns with sharp boundaries, governed by bistability and hysteresis. We further demonstrate how the hysteresis, position, timing, and precision of the boundary can be controlled, highlighting the dynamical flexibility of the circuit.
  • Cortical Travelling Waves: Mechanisms and Computational Principles

    Cortical Travelling Waves: Mechanisms and Computational Principles

    REVIEWS Cortical travelling waves: mechanisms and computational principles Lyle Muller1, Frédéric Chavane2, John Reynolds1 and Terrence J. Sejnowski1,3* Abstract | Multichannel recording technologies have revealed travelling waves of neural activity in multiple sensory, motor and cognitive systems. These waves can be spontaneously generated by recurrent circuits or evoked by external stimuli. They travel along brain networks at multiple scales, transiently modulating spiking and excitability as they pass. Here, we review recent experimental findings that have found evidence for travelling waves at single-area (mesoscopic) and whole-brain (macroscopic) scales. We place these findings in the context of the current theoretical understanding of wave generation and propagation in recurrent networks. During the large low-frequency rhythms of sleep or the relatively desynchronized state of the awake cortex, travelling waves may serve a variety of functions, from long-term memory consolidation to processing of dynamic visual stimuli. We explore new avenues for experimental and computational understanding of the role of spatiotemporal activity patterns in the cortex. Phase offsets Synchrony in the activity of neuronal populations has shape these waves so that they differ profoundly from 1,2 13 14 The differences in phase (an long been of interest in neuroscience , but its function sleep to wake . In turn, these waves transiently mod- amplitude-invariant measure of has remained elusive. Neural oscillations are rhythmic ulate neuronal excitability15–19 and shape responses to position in an oscillation cycle) fluctuations of spiking activity within neuronal popula- external input20 over a wide range of spatial and tempo- between two (or more) tions. It is proposed that synchrony of neural oscillations ral scales.
  • Unraveling the Distributed Neural Code of Facial Identity Through Spatiotemporal Pattern Analysis

    Unraveling the Distributed Neural Code of Facial Identity Through Spatiotemporal Pattern Analysis

    Unraveling the distributed neural code of facial identity through spatiotemporal pattern analysis Adrian Nestor1, David C. Plaut, and Marlene Behrmann Department of Psychology, Carnegie Mellon University, and Center for the Neural Basis of Cognition, Pittsburgh, PA 15213 Edited by Charles G. Gross, Princeton University, Princeton, NJ, and approved May 12, 2011 (received for review February 13, 2011) Face individuation is one of the most impressive achievements of Furthermore, insofar as face individuation is mediated by a our visual system, and yet uncovering the neural mechanisms network, it is important to determine how information is distrib- subserving this feat appears to elude traditional approaches to uted across the system. Some interesting clues come from the fact functional brain data analysis. The present study investigates the that right fusiform areas are sensitive to both low-level properties neural code of facial identity perception with the aim of ascertain- of faces (16, 25) and high-level factors (26, 27), suggesting that ing its distributed nature and informational basis. To this end, these areas may mediate between image-based and conceptual we use a sequence of multivariate pattern analyses applied to representations. If true, such an organization should be reflected in functional magnetic resonance imaging (fMRI) data. First, we the pattern of information sharing among different regions. combine information-based brain mapping and dynamic discrimi- The current work investigates the nature and the extent of nation
  • Spatio-Temporal Patterns Act As Computational Mechanisms Governing Emergent Behavior in Robotic Swarms

    Spatio-Temporal Patterns Act As Computational Mechanisms Governing Emergent Behavior in Robotic Swarms

    telligenc International Journal of Swarm n e I an rm d a E w v S o f l u o t l i o a n n a r r u Intelligence and Evolutionary y o C J l o a m n ISSN: 2090-4908p o i u t t a a n t r i o e t n n I Computation Theory Spatio-Temporal Patterns Act as Computational Mechanisms Governing Emergent Behavior in Robotic Swarms Terry Jack MDA*, Khuman SA and Owa K De Montfort University, Leicester, UK ABSTRACT Our goal is to control a robotic swarm without removing its swarm-like nature. In other words, we aim to intrinsically control a robotic swarm emergent behavior. Past attempts at governing robotic swarms or their self-coordinating emergent behavior, has proven ineffective, largely due to the swarm’s inherent randomness (making it difficult to predict) and utter simplicity (they lack a leader, any kind of centralized control, long-range communication, global knowledge, complex internal models and only operate on a couple of basic, reactive rules). The main problem is that emergent phenomena itself is not fully understood, despite being at the forefront of current research. Research into 1D and 2D Cellular Automata has uncovered a hidden computational layer which bridges the micro-macro gap (i.e., how individual behaviors at the micro-level influence the global behaviors on the macro-level). We hypothesize that there also lie embedded computational mechanisms at the heart of a robotic swarm’s emergent behavior. To test this theory, we proceeded to simulate robotic swarms (represented as both particles and dynamic networks) and then designed local rules to induce various types of intelligent, emergent behaviors (as well as designing genetic algorithms to evolve robotic swarms with emergent behaviors).
  • Spatiotemporal Pattern Formation in Neural fields with Linear Adaptation

    Spatiotemporal Pattern Formation in Neural fields with Linear Adaptation

    Spatiotemporal pattern formation in neural fields with linear adaptation Bard Ermentrout, Stefanos E. Folias, and Zachary P. Kilpatrick Abstract We study spatiotemporal patterns of activity that emerge in neural fields in the presence of linear adaptation. Using an amplitude equation approach, we show that bifurcations from the homogeneous rest state can lead to a wide variety of sta- tionary and propagating patterns, especially in the case of lateral-inhibitory synap- tic weights. Typical solutions are stationary bumps, traveling bumps, and stationary patterns. However, we do witness more exotic time-periodic patterns as well. Using linear stability analysis that perturbs about stationary and traveling bump solutions, we then study conditions for activity to lock to the position of an external input. This analysis is performed in both periodic and infinite one-dimensional spatial do- mains. Both Hopf and saddle-node bifurcations can signify the boundary beyond which stationary or traveling bumps fail to lock to external inputs. Just beyond Hopf bifurcations, bumps begin to oscillate, becoming breather or slosher solutions. 1 Introduction Neural fields that include local negative feedback have proven very useful in qualitatively describing the propagation of experimentally observed neural activ- ity [26, 39]. Disinhibited in vitro cortical slices can support traveling pulses and spiral waves [27, 53], suggesting that some process other than inhibition must cur- Bard Ermentrout University of Pittsburgh, Department of Mathematics, Pittsburgh PA e-mail: [email protected] Stefanos E. Folias University of Alaska Anchorage, Department of Mathematical Sciences, Anchorage AK e-mail: [email protected] Zachary P. Kilpatrick University of Houston, Department of Mathematics, Houston TX e-mail: [email protected] 1 2 Bard Ermentrout, Stefanos E.
  • What the Retinal Ganglion Cells Tell the Visual Cortex

    What the Retinal Ganglion Cells Tell the Visual Cortex

    LETTER Communicated by Michael Berry Rate Coding Versus Temporal Order Coding: What the Retinal Ganglion Cells Tell the Visual Cortex Rufin Van Rullen Simon J. Thorpe Centre de Recherche Cerveau et Cognition, FacultedeM´ edecine´ Rangueil, 31062 Toulouse Cedex, France It is often supposed that the messages sent to the visual cortex by the retinal ganglion cells are encoded by the mean firing rates observed on spike trains generated with a Poisson process. Using an information trans- mission approach, we evaluate the performances of two such codes, one based on the spike count and the other on the mean interspike interval, and compare the results with a rank order code, where the first ganglion cells to emit a spike are given a maximal weight. Our results show that the rate codes are far from optimal for fast information transmission and that the temporal structure of the spike train can be efficiently used to maximize the information transfer rate under conditions where each cell needs to fire only one spike. 1 Introduction How do neurons transmit information? This question is a central problem in the field of neuroscience (Perkel & Bullock, 1968). Signals can be conveyed by analog and electrical mechanisms locally, but over distances information has to be encoded in the spatiotemporal pattern of trains of action potentials generated by a population of neurons. The exact features of these spike trains that carry information between neurons need to be defined. The most commonly used code is one based on the firing rates of individual cells, but this is by no means the only option.