Neural Networks Towards Building a Neural Networks Community

Total Page:16

File Type:pdf, Size:1020Kb

Neural Networks Towards Building a Neural Networks Community Neural Networks 23 (2010) 1135–1138 Contents lists available at ScienceDirect Neural Networks journal homepage: www.elsevier.com/locate/neunet Editorial Towards building a neural networks community December 31, 2010 is my last day as the founding editor-in- Data about reinforcement learning in animals and about chief of Neural Networks and also my 71st birthday! Such vast cognitive dissonance in humans kindled a similar passion. What changes have occurred in our field since the first issue of Neural sort of space and time representations could describe both a Networks was published in 1987 that it seemed appropriate to thought and a feeling? How could thoughts and feelings interact my editorial colleagues for me to reflect on some of the changes to guide our decisions and actions to realize valued goals? Just as to which my own efforts contributed. Knowing where we as a in the case of serial learning, the non-occurrence of an event could community came from may, in addition to its intrinsic historical have dramatic consequences. For example, the non-occurrence of interest, be helpful towards clarifying where we hope to go and an expected reward can trigger reset of cognitive working memory, how to get there. My comments will necessarily be made from a emotional frustration, and exploratory motor activity to discover personal perspective. new ways to get the reward. What is an expectation? What is The founding of Neural Networks is intimately linked to the a reward? What type of dynamics could unify the description founding of the International Neural Network Society (INNS; of cognition, emotion, and action? Within the reinforcement http://www.inns.org) and of the International Joint Conference learning literature, there were also plenty of examples of irrational on Neural Networks (IJCNN; http://www.ijcnn2011.org). These behaviors. If evolution selects successful behaviors, then why are events, in turn, helped to trigger the formation of other groups of so many behaviors irrational? scientists and engineers who are interested in our field, as well as Such issues are still of current research interest. And all of them other conferences to which they could contribute to share their led me to the same theoretical method and modeling framework. ideas. The theoretical method, which I called the Method of Minimal There was little in the way of infrastructure in our community Anatomies, showed how analysis of how an individual's behavior thirty years ago. It was not obvious where to submit a modeling adapts autonomously in real time to a changing world can lead to article for publication, or what conferences to attend where one models of how the brain works. In other words, brains embody could report work to interested colleagues. It was also not clear a natural computational architecture for autonomous adaptation where one could go to school to learn how to model, or even to in real time to a changing world. By bridging the gap between learn about was known in the field. behavioral experience and brain dynamics, the method opened a How did I come to be in a position to contribute to these path towards solving the mind-body problem. developments? That happened in gradual stages over a period of This method led me to introduce networks of neurons in over fifty years. My own beginnings in our field started accidentally which short-term memory (STM), medium-term memory (MTM), in 1957, thirty years before Neural Networks was founded, when I and long-term memory (LTM) traces exist. The STM and LTM was a 17 year-old Freshman at Dartmouth College. At that time, like traces are also known as cell activities, or potentials, and adaptive thousands of other Freshmen, I took an introductory psychology course. The accident was my exposure to classical data about how weights. In such a network, backward effects in time became easy humans and animals learn about the world. To me, these data were to understand. These traces were described within Additive and filled with philosophical paradoxes. And 17 year-olds of a certain Shunting Models that lie at the core of essentially all contemporary disposition are particularly vulnerable to philosophical paradoxes! connectionist models of mind and brain. The Additive Model is For example, data about serial verbal learning in humans – now sometimes called the Hopfield model based on Hopfield's that is, the learning of lists of language or action items through popular 1984 article on this topic. When I introduced the model time – seemed to suggest that events can go ``backwards in time''. in 1957–58, both the paradigm and the equations were new, and Indeed, the non-occurrence of future items in a list to be learned even revolutionary. can dramatically change the distribution of performance errors at The LTM laws to which I was led included gated steepest all previously occurring list positions. The so-called ``bowed serial descent learning laws that allow learned increases and decreases position curve'' showed that the beginning and the end of a list in weights, in order to learn spatially distributed weight patterns. can be learned more easily than the middle, with error gradients From the outset, it was clear that the Hebb hypothesis that learned in performance that point forwards or backwards in time at the weights can only increase was incorrect. When in the 1970s von beginning or end of a list, and error gradients in the middle pointing der Malsburg and I introduced self-organizing maps, such a gated in both directions. In what sort of space and time could events steepest descent law was used in it, and Kohonen adapted this law go backwards in time? What type of dynamics could represent in his applications of self-organizing maps in the 1980s. the non-occurrence of an event? Exposure to such paradoxes The MTM law described how signals can be gated, or multiplied, triggered a passionate intellectual excitement that I had never by chemical transmitters or post-synaptic sites subject to activity- before experienced. dependent habituation. This law started to become broadly used 0893-6080/$ – see front matter ' 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.neunet.2010.10.006 1136 Editorial / Neural Networks 23 (2010) 1135–1138 in the 1990s due to the work of Abbott, Markram, and their scientific history to try to understand why it seemed so hard colleagues, often under the name of synaptic depression. for people to understand discoveries which, at least to me, Thus, while I was in college, it became clear that, to link mind seemed natural and intuitively appealing. I gradually began to and brain, one needed nonlinear neural networks operating on realize that I was contributing to a major scientific revolution multiple temporal and spatial scales. There was nothing like this whose groundwork was laid by great nineteenth physicists such going on in psychology departments then, which is why, with some as Hermann von Helmholtz, James Clerk Maxwell, and Ernst trepidation, I decided to get my Ph.D. in mathematics. When I Mach. These interdisciplinary scientists were physicists as well graduated from Dartmouth in 1961, the best research group in as psychologists and physiologists. Their discoveries made clear mathematical psychology, and one of the best departments in that understanding the Three N's of the brain would require applied mathematics, were at Stanford University. I therefore went new intuitions and mathematics. As a result, the next generation to Stanford to begin my graduate studies. While at Stanford, I tried of physicists exclusively pursued the major new paradigms of to get my nonlinear neural networks simulated with the help of one relativity theory and quantum theory, which could build on of the best programmers at Stanford's computation center, without great new physical intuitions that were supported by known success. mathematics. Psychologists and physiologists were left with Computers in those days were not up to the task of simulating inadequate intuitions and mathematics with which to explain nonlinear neural networks operating on multiple time scales. their data. Given inadequate tools with which to understand the The learning in such networks made them nonstationary in Three N's, a century of controversy, along with anti-theoretical time. The long-range interactions between neurons could be prejudices, unfolded in the mind-brain sciences as they collected said to be nonlocal. In other words, computers in those days huge, but typically unexplained, data bases. had problems simulating large networks of neurons undergoing Although these historical insights helped me to understand, and nonlinear, nonstationary, and nonlocal interactions—what I called emotionally cope with, the frightening social forces to which I was the Three N's of brain dynamics. My mathematical training helped exposed as a young man, it was still difficult to deal with them. Had me to overcome this roadblock by giving me enough tools to prove I not been first in my class in school for many years, and were it not foundational theorems in the 1960s and 1970s about how STM, a time when the United States was investing large sums of money MTM, and LTM interact, including the first global theorems about in student fellowships, I may not have survived the social pressures content addressable neural memories, and theorems about how against the kind of research that I was doing. the bowed serial position curve arises. This loneliness, combined with my passionate belief in the The theoretical method also facilitated technology transfer importance of this new paradigm shift, drove me to do everything to applications in neuromorphic engineering and technology. that I could, when opportunities presented themselves, to build It did this by clarifying how brain mechanisms give rise to a community wherein future students and practicing scientists behavioral functions while an individual autonomously adapts could readily learn about our field, as if it were the most natural to a changing world.
Recommended publications
  • Adaptive Resonance Theory
    Adaptive Resonance Theory Jonas Jacobi, Felix J. Oppermann C.v.O. Universitat¨ Oldenburg Gliederung 1. Neuronale Netze 2. Stabilität - Plastizität 3. ART-1 4. ART-2 5. ARTMAP 6. Anwendung 7. Zusammenfassung Adaptive Resonance Theory – p.1/27 Neuronale Netze Künstliche neuronale Netze sind Modell für das Nervensystem Aufbau aus einzellnen Zellen Paralelle Datenverarbeitung Lernen implizit durch Gewichtsveränderung Sehr unterschiedliche Strukturen Adaptive Resonance Theory – p.2/27 Neuronale Netze Summation der Eingänge Vergleich mit einem Schwellenwert Erzeugen eines Ausgabesignals Adaptive Resonance Theory – p.3/27 Stabilität $ Plastizität Lernen mit fester Lernrate Die Lernrate gibt an wie stark ein neues Muster die Gewichte beeinflusst Lernen erfolgt bei überwachtem Lernen nur in einer Trainingsphase Im biologischen System wird das Lernen nicht beendet Meist kein direktes Fehlersignal Adaptive Resonance Theory – p.4/27 Stabilität $ Plastizität Stabilität Einmal Gelerntes soll gespeichert bleiben Das trainieren neuer Muster soll vermieden werden, da diese die Klassifikation stark verändern könnten Plastizität Neue Muster sollen kein vollständiges Neutrainieren erfordern Alte Klassifikation sollten beim Hinzufügen neuer Muster erhalten bleiben Adaptive Resonance Theory – p.5/27 Stabilität $ Plastizität Beide Anforderungen konkurrieren Hohe Stabilität oder hohe Plastizität Plastizitäts-Stabilitäts-Dilemma Adaptive Resonance Theory – p.6/27 Stabilität $ Plastizität Lösungsansätze: Kontrolle ob das neue Muster bereits bekannt ist Für neue Muster eine neue Klasse öffnen Bei hoher Ähnlichkeit zu bekannten Mustern Einordnung in diese Klasse Der Parameter der Kontrolle kann als Aufmerksamkeit betrachtet werden Adaptive Resonance Theory – p.7/27 Stephen Grossberg Stuyvesant High School, Manhattan, 1957 Rockefeller University, Ph.D., 1964-1967 Associate Professor of Applied Mathematics, M.I.T., 1969-1975.
    [Show full text]
  • Stephen Grossberg
    STEPHEN GROSSBERG Wang Professor of Cognitive and Neural Systems Professor of Mathematics, Psychology, and Biomedical Engineering Director, Center for Adaptive Systems Boston University 677 Beacon Street Boston, MA 02215 (617) 353-7857 (617) 353-7755 [email protected] http://cns.bu.edu/~steve HIGH SCHOOL: Stuyvesant High School, Manhattan First in Class of 1957 COLLEGE: Dartmouth College, B.A. First in Class of 1961 A.P. Sloan National Scholar Phi Beta Kappa Prize NSF Undergraduate Research Fellow GRADUATE WORK: Stanford University, M.S., 1961-1964 NSF Graduate Fellowship Woodrow Wilson Graduate Fellowship Rockefeller University, Ph.D., 1964-1967 Rockefeller University Graduate Fellowship POST-GRADUATE ACTIVITIES: 1. Assistant Professor of Applied Mathematics, M.I.T., 1967-1969. 2. Senior Visiting Fellow of the Science Research Council of England, 1969. 3. Norbert Wiener Medal for Cybernetics, 1969. 4. A.P. Sloan Research Fellow, 1969-1971. 5. Associate Professor of Applied Mathematics, M.I.T., 1969-1975. 1 6. Professor of Mathematics, Psychology, and Biomedical Engineering, Boston University, 1975-. 7. Invited lectures in Australia, Austria, Belgium, Bulgaria, Canada, Denmark, England, Finland, France, Germany, Greece, Hong Kong, Israel, Italy, Japan, The Netherlands, Norway, Qatar, Scotland, Singapore, Spain, Sweden, Switzerland, and throughout the United States. 8. Editor of the journals Adaptive Behavior; Applied Intelligence; Behavioral and Brain Sciences (Associate Editor for Computational Neuroscience); Autism Open Access Journal; Behavioural
    [Show full text]
  • Otices of The
    OTICES OF THE AMERICAN MATHEMATICAL SOCIETY 1991 AMS-MAA Survey First Report page 1086 NOVEMBER 1991, VOLUME 38, NUMBER 9 Providence, Rhode Island, USA ISSN 0002-9920 Calendar of AMS Meetings and Conferences This calendar lists all meetings approved prior to the date this issue went to press. is possible. Abstracts should be submitted on special forms which are available The summer and annual meetings are joint meetings of the Mathematical Asso­ in many departments of mathematics and from the headquarters office of the So­ ciation of America and the American Mathematical Society. The meeting dates ciety. Abstracts of papers to be presented at the meeting must be received at the which fall rather far in the future are subject to change; this is particularly true headquarters of the Society in Providence, Rhode Island, on or before the deadline of meetings to which no numbers have been assigned. Programs of the meet­ given below for the meeting. The abstract deadlines listed below should be care­ ings will appear in the issues indicated below. First and supplementary announce­ fully reviewed since an abstract deadline may expire before publication of a first ments of the meetings will have appeared in earlier issues. Abstracts of papers announcement. Note that the deadline for abstracts for consideration for presenta­ presented at a meeting of the Society are published in the journal Abstracts of tion at special sessions is usually three weeks earlier than that specified below. For papers presented to the American Mathematical Society in the issue correspond­ additional information, consult the meeting announcements and the list of special ing to that of the Notices which contains the program of the meeting, insofar as sessions.
    [Show full text]
  • Knowledge Inn (In Nature)
    Available online at www.joac.info ISSN: 2278-1862 Journal of Applicable Chemistry 2018, 7 (6): 1843-1848 (International Peer Reviewed Journal) Knowledge Inn (in nature) Toxicological Knowledge Discovery Knowledge Extraction in Chemical kinetics KLab rsr.chem1979 Research Profile of Stephen Grossberg, Boston Toxicological Knowledge Discovery Information Source (is) ACS.org ; Richard Sherhod et al, J. Chem. Inf. Model. 2014, 54, 1864−1879 Emerging pattern mining to aid toxicological knowledge discovery; doi.org/10.1021/ci5001828 | AAA Knowledge Inn 1843 Knowledge Extraction in Chemical kinetics Information Source (is) : ACS.org ; TI: Extracting Knowledge from Data through Catalysis Informatics AU: Andrew J. Medford, M. Ross Kunz, Sarah M. Ewing, Tammie Borders, Rebecca Fushimi JO: ACS Catal. 2018, 8, 7403−7429 DOI: 10.1021/acscatal.8b01708 Schematic of data−information−knowledge hierarchy Schematic relationship between in heterogeneous catalysis (data Inference/information knowledge/Intelligence (Diki) o Data driven/ model driven Models (empirical, theoretical, computational, simulation based on statistics and/or physics) o Creation of derived data, establish heuristic relationships between data, rapid estimation of parameters for reaction network or atomic-scale surface models o High-dimensional structures based on reaction network/atomic-scale structure distilled reaction mechanism(s) and active site(s) [Design of experiments; refinement of micro-kinetic models based on chemical master equation] AAA Knowledge Inn 1844 Research
    [Show full text]
  • Stephen Grossberg
    Talking Nets: An Oral History of Neural Networks edited by James A. Anderson and Edward Rosenfeld FirstMIT Presspaperback edition , 2(XK) (C> 1998Massachusetts Institute of Technology AU rights reserved. No part of this book may be reproducedin any fonD by any electronic or mechanicalmeans (including photocopying , recording, or informationstorage and retrieval ) without permissionin writing from the publisher. " " This book was set in Palatinoon the Monotype Prism Plus PostScriptimage setterby Asco TradeTypeset ting Ltd., HongKong , andwas printed and bound in the UnitedStates of America. Photographsof Gail Carpenterand Stephen Gros ,bergby DeborahGrossberg . Photographsof JamesA Andenonby PhilipLiebennan . All otherphotograph! by JamesA Anderson. Ubraryof CongressCatalo~ -in-PublicationData Talkingnets : an oral historyof neuralnetworks / editedby JamesA Andersonand Edward Rosenfeld p. an. "A Bradfordbook ." Includesbibliographical rpJ~ ~ and ind~x- ISBN0 -262-01167-0 (hardcover: alk. paper), 0-262-51111-8 (pb) 1. Neuralcomputers . 2. Neuralnetworks (Computer science ) 3. Scientists- Interviews. I. Anderson. TamesA . II. Rosenfeld. Edward. QA76.87.T37 1998 006.3'2' O922- dc21 97-23868 . TalkingAndersonNets 8 StephenGrossberg StephenGrossberg is Wang Professorof Cognitiveand Neural Systems; Professorof Mathematics , Psychology, and BiomedicalEngineering ; Chairman, Depari1nentof Cognitiveand Neural Systemsand Director, Centerfor Adaptive Systems, Boston , Boston, University' Massachusetts. A good to read about s work is a 1995 " place" further ProfessorGrossberg paper, The Attentive Brain, American Scientist 83: 438- 449. July 1993 , Portland , Oregon ER: What is your date of birth and where were you born ? SG: December 31, 1939, in New York City . ER: Tell us something about your parents and what your early childhood was like . ' SG: My grandparents were &om Hungary , around Budapest, so I m a - second generation American .
    [Show full text]
  • Grossberg, CV 8-14-21
    STEPHEN GROSSBERG Wang Professor of Cognitive and Neural Systems Emeritus Professor of Mathematics & Statistics, Psychological & Brain Sciences, and Biomedical Engineering Director, Center for Adaptive Systems Boston University 677 Beacon Street Boston, MA 02215 [email protected] sites.bu.edu/steveg HIGH SCHOOL: Stuyvesant High School, Manhattan First in Class of 1957 COLLEGE: Dartmouth College, B.A. First in Class of 1961 A.P. Sloan National Scholar Phi Beta Kappa Prize NSF Undergraduate Research Fellow GRADUATE WORK: Stanford University, M.S., 1961-1964 NSF Graduate Fellowship Woodrow Wilson Graduate Fellowship Rockefeller University, Ph.D., 1964-1967 Rockefeller University Graduate Fellowship POST-GRADUATE ACTIVITIES: 1. Assistant Professor of Applied Mathematics, M.I.T., 1967-1969. 2. Senior Visiting Fellow of the Science Research Council of England, 1969. 3. Norbert Wiener Medal for Cybernetics, 1969. 4. A.P. Sloan Research Fellow, 1969-1971. 5. Associate Professor of Applied Mathematics, M.I.T., 1969-1975. 6. Professor of Mathematics, Psychology, and Biomedical Engineering, Boston University, 1975-. 1 7. Invited lectures in Australia, Austria, Belgium, Bulgaria, Canada, China, Denmark, England, Finland, France, Germany, Greece, Hong Kong, Israel, Italy, Japan, The Netherlands, Norway, Qatar, Scotland, Singapore, Spain, Sweden, Switzerland, and throughout the United States. 8. Editor of the journals Adaptive Behavior; Applied Intelligence; Behavioral and Brain Sciences (Associate Editor for Computational Neuroscience); Autism Open Access
    [Show full text]
  • February 1980 Table of Contents
    •ces Mathematical Society < 2. c ~ ..N..... zc 3 lf N CALENDAR OF AMS MEETINGS THIS CALENDAR lists all meetings which have been approved by the Council prior to the date this issue of the Notices was sent to press. The summer and annual meetings are joint meetings of the Mathematical Association of America and the Ameri· can Mathematical Society. The meeting dates which fall rather far in the future are subject to change; this is particularly true of meetings to which no numbers have yet been assigned. Programs of the meetings will appear in the issues indicated below. Pirst and second announcements of the meetings will have appeared in earlier issues. ABSTRACTS OF PAPERS presented at a meeting of the Society are published in the journal Abstracts of papers presented to the American Mathematical Society in the issue corresponding to that of the Notices which contains the program of the meet· ing. Abstracts should be submitted on special forms which are available in many departments of mathematics and from the office of the Society in Providence. Abstracts of papers to be presented at the meeting must be received at the headquarters of the Society in Providence, Rhode Island, on or before the deadline given below for the meeting. Note that the deadline for ab· stracts submitted for consideration for presentation at special sessions is usually three weeks earlier than that specified below. For additional information consult the meeting announcement and the list of organizers of special sessions. MEETING ABSTRACT NUMBER DATE PLACE DEADLINE
    [Show full text]
  • Natural Intelligence
    ISSN 2164-8522 Natural Intelligence The INNS Magazine Volume 1, Issue 2, Winter 2012 NI Can Do Better Compressive Sensing Evolving, Probabilistic Spiking Neural Networks and Neurogenetic Systems Biologically Motivated Selective Attention CASA: Biologically Inspired approaches for Auditory Scene Analysis Retail: US$10.00 Natural Intelligence: the INNS Magazine Editorial Board Editor-in-Chief Soo-Young Lee, Korea [email protected] Associate Editors Wlodek Duch, Poland [email protected] Marco Gori, Italy [email protected] Nik Kasabov, New Zealand [email protected] Irwin King, China [email protected] Robert Kozma, US [email protected] Minho Lee, Korea [email protected] Francesco Carlo Morabito, Italy [email protected] Klaus Obermayer, Germany [email protected] Ron Sun, US [email protected] Natural Intelligence: the INNS Magazine 2 Volume 1, Issue 2, Winter 2012 ISSN 2164-8522 Natural Intelligence The INNS Magazine Volume 1, Issue 2, Winter 2012 Regular Papers 5 NI Can Do Better Compressive Sensing by Harold Szu 23 Evolving, Probabilistic Spiking Neural Networks and Neurogenetic Systems for Spatio- and Spectro- Temporal Data Modelling and Pattern Recognition by Nikola Kasabov 38 Biologically Motivated Selective Attention Model by Minho Lee 50 CASA: Biologically Inspired approaches for Auditory Scene Analysis by Azam Rabiee, Saeed Setayeshi, and Soo-Young Lee Columns 4 Message from the Vice President for Membership News 59 2012 INNS Awards 64 2012 INNS New Members at the Board of Governors Call for Papers 65 Call for Abstracts on Neuroscience & Neurocognition, IJCNN2012, 2012 International Joint Conference on Neural Networks, Brisbane, Australia 65 WIRN 2012, 22nd Italian Workshop on Neural Networks, Salerno, Italy Volume 1, Issue.2, Winter 2012 3 Natural Intelligence: the INNS Magazine A Message from the INNS Vice President for Membership New Excitements in 2012 Irwin King INNS Vice-President for Membership As the Vice-President for Membership, I would like to update you on a few exciting and important items related to INNS members.
    [Show full text]