International Centre for Theoretical Physics

No. 24125July/August 1989

ICTP Dirac Medalsand Yukawa Prize

Professor Efim Samoilovich Fradkin,winner of the ICTP Dirac Medal 1988,was awarded the prize at ICTP on 11July 1989. The citation reads: "for hismany fruitful contributions to thedevelopment of quantum field theory andstatistics. Among these are his earlywork on functional methods includinghis formal solution to the Schwinger-Dyson equations for the Green'sfunctions of interacting systems. Thisresult hss become a standard part ofmodern quantum field theory.Independently ofTakahnshi he discoveredthe generalized Ward identitites forelectrodynamics. These identities andtheir generalization for non-Abeliangauge theories are basic to theunderstanding of local symmetries. Inhis work on the Schwinger-Dysonequations, Fradkin drew attention to thezero-charge problem, a potentialinconsistency in Abelian gauge theories

whose later resolution in the non-Abelian theories led to the discovery ofasymptotic freedom. At the same timeas Schwinger and Nakano, Fradkinconstructed a Euclidean formulation ofquantum field theory, a developmentwhich was to have far-reachingimplications for the development ofstatistical physics and string theory. IIiscontributions to the quantization ofrelativistic systetns with constraints arewidely recongized. This workculminated in the Baalin-Fradkin-Wilkovisky quantization method whichis used both in quantum field theory andin the theory of extended objects such asstrings and membranes".

The other 1988 Dirac Medal wasawarded to Prof. David Gross @rincetonUniversity, New Jersey, USA) on 10April 1989.

Michael B. Creen from Queen MaryCollege, London, UK, and John H.Schwarz from the California Institute ofTechnology, Pasadena, USA, are therecipients of the Dirac Medals of thc

Professor Abdus Salam, Director af ICTP and President of TWAS, reading theiitation for; the 198 Dirac Medal a'warded to Prcf. Efim S. Fradkin. From left toright: Prof. H. Dalafi, Executive Secretary of the World Laboratory Branch inTrieste, Prof. A. Zichichi, President of the World Laboratory and Member of theScientific Council of the ICTP, Prof. Abdus Salarn, Director of the ICTP, andPresidert of the Third World Academy of Sciences, and Prof. E.S. Fradkin (USSR),ICTP Dirac Medal 1988.

International Centre for TheoreticalPhysics 1989, "for their basiccontributions to the development ofsuperstring theory. Most significantwas their discovery that chiral gaugeanomalies are absent for a class of ten;

dimensional superstring theories. Thisprovided a strong indication thatsuperstring theory with a specific gaugesymmetry may provide a consistentunified quentum theory of thefundamental forces including gravity. Itled to an explosion of interest in stringtheory which has already spurredremarkable cdvances both inmathematical physics and in puremathematics".

The Medals will be presented to theawardees later on.

ContentsICTP Dirac Medals

and Yukawa Prize I

Abdus Salam Honorary Knight )

25th Anniversary of t\g ICTP J

Frontiers of Informatics:Artificial Intelligenceand Parallel Machines 3Nerv Centre for TheoreticalResearch in Bangalore 5

Book review from IAEA 5

Activities at ICTPJu.ly-August 1989 6

Future Activities at ICTP 8

This was announced, as every yearsince 1985, on 8 August, birthday of thelate Paul Adrien Maurice Dirac - NobelPrize for Physics 1933 - on the sameday of thc award ceremony of the ICTPPrize in honout, this year, of the lateHideki Yukawa, the great Japanese

Nervs from ICTP - No. 24125 - July/August 1989

physicist who won the Nobel Prize in7949. Dr. Ashoke Sen from the TataInstitute of Fundamental Research,Bombay, India, is the recipient of theYukawa Prize 'for his contributions tostring theory, and in particular for theapplication of the sigma model approachto the heterotic string theory. IIe hascontributed to the development oftechniques of superstring perturbationtheory at higher genus which haveenhanced the string theory understandingof space-time supersymmetry non-renorm.alization theorems. He has alsostudied the implications of modularinvariance in the problem of theclassification of rational conformal fieldtheories in 2 dimensions".

The Ceremony took place in theMain Lecture Hall of the ICTP.Professor Abdus Salam, Dircctor of theInternational Centre for TheoreticalPhysics, read the citation for bothawards, and presenfed the medal, diplomaand US$ 1,000 cheque for the YukawaPrize.

Two ICTP Medals were alsopresented to Mr. S. Coloni, Member ofthe Italian Parliament, and Mr. M.Rossetti, Member of the EuropeanParliament, in appreciation of thet workfor tlre progress of science in Trieste.

_ * _The Dirac Medals of the International

Centre for Theoretical Physics wereinstituted in 1985 in memory of Prof.P.A.M. Dirac, a honoured gu.est at and astaunch friend of the ICTP. The Mqdalsare awarded yearly for conlributions totheoretical physics.

From left to right: Dr. Ashokc Sen, awardee of the llideki Yukawa Prize,Professor Abdus Salam and Prof. P. Budinich (International School for AdvancedStudies).

The 1985 Dirac Medals were awardedto Professor Yakov Zeldovich (Institutefor Space Research, Moscow, USSR)and Prof. Edward Witten (PrincetonUniversity, USA) and in 1986 to Prof.Yoichiro Nambu (Enrico Fermi Institutefor Nuclear Studies, Chicago Universiry,USA) and Prof. Alexander Polyakov(Landau Institute for TheoreticalPhysics, Moscow, USSR). In 1987,they were awarded to Prof. Bryce DeWitt(University of Texas at Austin, USA)and Prof. Bruno Zumino (University ofCalifurnia at Berkeley, USA). Therecipients of the 19BB Medals were Prof.David J. Gross (Princeton University,

Prof. A. Zichichi presenting the Dirac Medal to Prof. E.S. Fradkin.

USA) and Prof. Efim S. Fradkin(Lebedev Physical Institute, Moscow,ussn).

The Selection Committee includesProfessors S. Lundqvist, R. Marshak, J.Schwinger, S. Weinberg, Abdus Salamand others. The Dirac Medals of theICW are not awarded to Nobel Laureatesor Wolf Foundation Prize winners.

_ * _One ICTP Prize is awarded yearly for

outstanding and original contributionswithin an announced field, to a scientistunder 40 years of age, national of adeveloping country, working and livingin a developing country. The winner ofthe Prize is selected by qn InternationalCommittee, from among the mostoutstanding scientists in the announcedfield.

Abdus SalamHonorary Knight

Courtesy ofThe Times

Thur, 3 Aug 1989

Professor Abdus Salam, professor oftheoretical physics at Imperial College,London, and the inspiration behind thefoundation of tre International Centre forTheoretical Physics at Trieste, has beenappointed an honorary KnightCommander of the Order of the BritishEmpire.

The scientists and staff of the ICTPjoin in congratulating Professor Abdus

News from ICTP - No. 24125 - July/August 1989

Salam on his appointment.

25th Anniversarvof the ICTP

A quarter of a century has passedsince Abdus Salam opened the Seminaron Plasma Physics, the inauguralactivity of the ICTP which had jusr beencreated. Since r,hen, from US$350,000,the annual budget has gone rc 2Amillion, 28,000 scientists fromdeveloping countries have passedthrough Trieste out of a total of 40,000,there are about forty-five differentactivities each year, the Centre nowoccupies four large buildings, and 4,300preprints have been produced.

All these achievements are due to thefar-sighted and energetic guidance ofAbdus Salam, Nobel Laureate forPhysics 1979 and Director since theICTP's inception, the patronage of theIAEA and UNESCO, the generouscontribution from the Government ofItaly and of many others. The 25thAnniversary will be celebrated on 3lOctober 1989. The ceremony will behonoured by the presence of thePresident of Inly, Mr. F. Cossiga, andof many other distinguished guests fromthe international, scientific and poliricalworld. The official part of the ceremonywill be followed by a ttrree-day scientificmeeting.

Frontiers of Informatics:Artificial Intelligenceand Parallel Machines

byTahir Shah

Courtesy ofTWAS Newsletter

The origin of informatics lies deepwithin the abstract studies of symboliclogic and the foundations ofmathematics. David Hilbert, one of themost versatile and bri l l iantmathematicians of this century,formulated a proof theory in which thetruth of mathematical formulae isdecidible mechanically by a finirenumber of steps. This was a remarkableidea. The decision theoretical part of hisformalism, dre famous "Entscheidungs-problem", led the British mathematicianAllan Turing to devise a model ofcomputing machines. All computers are

based on the Turing-machines model.The proof theorctical part of Hilbert'sprogram led to the foundations ofArtificial Intelligence. Two most widelyused symbolic computer languages,LISP and PROLOG, owe their existenceto two sub-branches of symbolic logic— Church's lambda calculus and HornClause logic, respectively. Without suchtheoretical studies perhaps even ttre birthof information technology would nothave been possible.

ktificial Intelligence has progressedfrom a laboratory science to anexpanding technology over the last fewyears. Since the announcement by theJapanese of their Fifth Generationcomputer project in 1981, there has beena plethora of activity in this field fromall corners of the globe. In his recentbook "Infinite in All Directions,,,theoretical physicist Freeman J. Dysonnamed three technologies which willrevolutionize our future. These are(a) genetic engineering, (b) artificialintelligcnce, and (c) space science.According to him, "The secondtechnological revolution is artificialintelligence. This revolution has alreadybegun with the rapid development andproliferation of computers..." However,despite mass-media attention, there is aconfusion as to the nature of A.I., evenamong scientists who are not specialistsor actively involved in A.I.

Quite often "expert systems" areconsidered synonymous with A.I. Inother cases, it is confused with the studyof neural models only.

One of the points lhat requiresclarification at the outset, therefore, isthe commonly conceived myth thatexpert systcms constitute the toml scopeof Artificial Intelligence. Experr systemsare not all there is to A.I. and rule-basedsystems are not all there is to the field ofexpen systems.

Devised in thc mid-70s, rule-basedexpert systems were one of the firstcommercially developed packages. Dueto the availability of comparatively weakcomputing power at that time, it isunderstandable ftat expert systems weredesigned using the simplest knowledgerepresentation schemes. So-calledproduct"ion rules are very similar to theif-then statements being used forprocedural programming languages.Procedural programming languages usealgorithmic procedures to solve aproblem. Most of the well-knowncommercially used languages (e.g.Cobol, Fortran and Pascal) are of this

vanety.In the field of A.I., however, one

rarely uses procedural languages becausethey are inappropriate for symbolic anddescriptive programming style. Toillustrate the complexity andinterdisciplinary nature of A.I., sometechnical details of an intelligent systemwould not be out of context here.

Intel l igenceOver the centuries, philosophers and

other intellectuals have tried to defineintelligence. Intuitively, it is the abilityto solve problems, learn fromexperience, infer facts from other factsand so on. In computational studies ofintelligence we take for granted thatmachine intelligence must deal withthese capabiliries. Complex inrelligentbehaviour involves the synergy of anumber of minor subsystems.

Over the last two decades, researchinto A.I. and related cognitive scienceshas led to the conclusion thatcomputational studies of intelligencedgal with two basic problems: (1). Howto represent knowledge, i.e. the study ofknowlcdge representation schemes; (2)How to manipulate knowledge throughinferential techniques such as deduction,induction, property inheritance, etc.(Inference is a form of knowledgemanipulation relating one piece ofknowledge wittr another).

Intelligence itself, more specifically,is defined as a problem-solving skill inthe broadest sense imaginable andincludes speech, natural language andother perceptual processes such asreasoning, abstraction, acquisition andmanipulation of knowledge. Thus, tomake a machine intelligent, we need torepresent acquired knowledge in aconvenient, concise and consistent form.This study of representation is a centralproblem in the field of A.I. research. Thedifficulty is in developing sufficientlyprecise nolation with which to representknowledge. A specitic descriprion ofknowledge is called a knowledgerepresentation scheme (KRS).

For practical purposes, we canconsider a knowlcdge base as a model ofthe physical world. Typically, the worldaround us is made up of a collection ofobjects. The collection of all of theseobjects and their relationships to oneanolher at any moment in timeconstil.ute a state. There can also be statetransformations resulting in creation,destruction or change in theserelationships.

Nerls from ICTP - \ o . 24125 - July/August 1989

There are many well-establishedknowledge representation schemes,however, these four are used most often:

Logic: Symbolic logic dates back toancient Indian and Greek philosophers. Itwas only at the end of the 19th centurythat a formal approach was started byGeorge Boole ftnown for his Booleanlogic, used in the field of electricalengineering), Gottlob Frege, BertrandRussell and others. Today, one of themost promising A.I. languages, calledPROLOG (PROgramming in LOGic) isbased on symbolic logic. In this KRS,true assertions about states are stored ina knowledge base.

Production Rules: This is the mostwidely used representation scheme byindustrial/commercial expert systemdesigners. The knowledge base consistsof a set of if...then rules, i.e. "if acondition is true, then activate aprocedure". In this scheme, the rules areactually state transformations betweenworld states.

Frames: In this representationscheme, facts ate represented ascontextual structures with slots whichare to be filled for details, For instance,identical names can fill different frameswithout confusion, because the systemdoes not confuse different meanings ofttre same word in different contexts.

Semantic Net: This is a network ofconceptual objects connected to eachother through relations. A semantic netis a collection of individuals andrelationships arranged in a suitablenetwork.

InferenceThe inference engine is the part of an

intelligent system that relates one pieceof knowledge (assertion or fact) toanother. It is the component of an expertsystem that controls its operation byselecting the rules to use, accessing andexecuting those rules, and determiningwhen an acceptable solution has beenfound. This component is sometimescalled the control structure or the ruleinterpreter.

Expert systems are therefore specialtypes of knowledge-based systems. Inthese systems, an gxpert's knowledgefrom a narrow domain is stored in thekdowledge base. Typically, acqmmercial expert system is designedusing production rules and is targeted ateither diagnostic or decision supporttasks.

In applying A.I. to the field ofrobotics, we find that although the

problem robots encountcr are simple andintuitive, they do represent majordifficulties in the plan generationdomain. A plan is an essential steptowards solving any problem. A simpledefinition of a plan would be an orderedsequence of acf.ions to reach the solutionof a problcm. A plan can also be definedas the representation of a course ofaction listing sub-goals in some orderedfashion. There are four distinct planningapproaches: hierarchical, non-hierarchical, script-based andopportunisL

A hierarchical planner, for example,generates a hierarchy of planrepresentations wherein the highest is asimplification of the plan and the lowestis a dctailed plan. The robot controller'sknowledge would contain commands forits mechanical motion, the perception ofthe woild around it, the formulation ofan action plan and the monitoring andexecution of that plan.

One project on the leading edge ofrobotics and A.I. is in space-relatedautomation projects, specificallyNASA's Space Station design anddevelopment. The applications in thisarea include: teleoperation and roboticsfor physical objecr manipulation; expertsystems to aid in monitoring, diagnosisand maintenance; planning, to schedulespace station resources and to determinewhat action should be performed byautonomous robots; dala managementsystems; and man-machine interfaces.

With the further addition ofintelligence, autonomous robots withself-contained vision, planning andcontrol, will be able to perceive andmanipulate objects. With suchcapabiliries, the robots will be able tomove around within the space station tocarry out the crew's orders. Applicationsin the manufacturing arena on eailhwould be forthcoming as a result ofthese innovations. All this is expected tohappen by the year 2005.

On the expert system side, theultimate objective is to have a spacemanufacturing system that is capable ofquality control, process control andmaintenance. This is to be achieved inthree phases. The first during 1991-1992, rhe second L993-1995, and thethird from 2000-2005.

On the planning side, robots areexpected to function very much likehuman beings. By the end of 2005, it isthought that these robots, all of themindependent and autonomous, will movearound from one location to another and

work cooperatively with other robots.They will be able to achieve such tasksas planning complex maintenance andrepair operations for manufacturingequipment and subsystems.

Massively parallel machinesIn October 1987,I made a projection

about parallel computers published inCanadian Electronic EngineeringBebruary 19881 that by the mid-ninetieswe expect to see a widespread use ofparallel machines in image processing,graphics, and some large CAD systems.Towards the end of the next decade,micro and mini size parallel machineswill be as popular as we seemicrocomputers loday. A recent issue ofByte magazine (USA) dedicated topersonal parallel machines suggests thatmy prediction was correct.

In the late seventies, when LISPmachines entered tle computer marketand later on, when the Japanese FifthGeneration computer project wasannounced, most felt that specializedmachines like LISP machines and "to bedeveloped inference machines" woulddominate the market of the late eightiesand onwards. This was due to the factthat there were great expectations fromthe newly publicized ArtificialIntelligence. Despite the apparentpopularity of Artificial Intelligence inrecent years, specialized machines likeLISP or Inference machines are usedonly for either rapid proto-typing orresearch and development. The generalpurpose single CPU machine still is thework horse of a large segment of thecomputing industry.

For some sectors of computing, suchas image and radar sigrial processing,complex fluid dynamics andenvironmental model calculations andother similar demanding applications,the successful use of parallel machinesindicates rheir feasibility and eventuallarge scale usage. In fact, for manyproblems, a single processor is nolonger good enough. For example,consider an image processing systemwhich can recognize and describe apicture. An ordinary television picturewith 1000 x 1000 'pixels (pictureelements) may take a processing time of10 to 20 hours on such machines asDEC PDP-10 (where each pixel is an 8-bit-element). In comparison, a parallelmachine, such as MPP (MassivelyPnrallel Processor) built by GoodyearAerospace for NASA, can process7123.5 million pixels per second for

News from ICTP - No. 2'{/25 - July/August 1989

gray scale thresholding. The gray scale isa measure of blackness for a black andwhite picture. This machine contains16,384 processors interconnected in a128 x 128 configuration with a basiccycle rate of 100 nano-seconds.

There is a demand for more and moreprocessing power as new kinds ofsoftware systems, such as naturallanguage processing, are coming intogreater use for industrial and commercialapplications. Although the price for asingle VLSI processor is decreasing,single processor machines do not satisfymany large scale applicationrequirements. Some experimentalsystems were built recently to take careof this need.

One would typically extrapolate threefactors to determine the future trend.These are: (a) economic feasibility, (b)limits on existing technology, and (c)scientific breakthroughs. To analyzesystematically the influence of thesethree factors, the hardware technologiescan be classified in terms of distinctareas for improvement, namely:

Conventional electronics: (a) VSLIimprovements based on productiontechnologies, such as surface mountcomponents, gate arrays, programmablearray logic, etc.; O) New semiconduclorssuch as GaAs; (c) Specialized chips suchas the LISP machine chip.

New physical principles: (a) Opticalgates and eventually optical VLSI;(b) Bio-chemical chips.

New design pinciples: (a) Innovativenon Turing-Von Neumann machines; (b)Data flow machines; (c) Fixed andvariable topology parallel machines.

Parallel machines are in the lattercategory because these machines can bebuilt using either conventional or exotictechnologies, so let us see why parallelmachines are going to dominate ourfuture.

Parallel dominanceSince the first LSI chip in the early

seventies, the density of active elements,i.e. number of the transistors per squarecentimetre, is increasing continuously.More and more transistors are beingsqueezed into a smaller and smaller chip.In 1959 only one transistor would fit ona ciip and in 1987 the number surpasseda million. But there are limitations onsuch increases. The internal circuits of achip are wired using a very fine line ofconducting material and are referred to asthe "lines". These lines are very thin,almost one millionth of a centimetre or

less wide. The line width has reachcdnow almost near molecular dimensionsand the laws of physics suggest that wecan not nzurow it down significantly anyfurther. There are othsr reasons besidcsthis, however. New processing andlithographic techniques may make itpossible in the ncar future tomanufacture VLSI with ten or a hundredmillion transistors. The argumentagainst such an unbounded increase inthe size and the density of VLSI is thatmere possibilities inherent in large scaleintegration do not tlemselves createadequate ways of exploiting thistechnology efficiently.

Hardware alternativesThe other hardware altcrnatives are to

build optical processing elements ormove into biochip manufacturing. Boththese technologies are ncither mature noreconomically feasible at the presenttime. So we are still left with good oldelectronics and VLSI — cheap andfunctional. Also, one does not expectthat any new discovery will expeditedevelopment of optical or biochemicalgates. In both cases the physicalprinciples are well understood. However,the problem is not the physics orchemistry behind the idea but rhematurity of the technology which makesit economically feasible. This is what islacking in optical and biochemical chipdevelopment. It takes a full two decadesto mature a technology from its infancyto a robust and usable stage.Conclusively, the natural directionwould be to develop parallel machinesfurther.

The other faclors are of a theoreticalnature but nevertheless of fundamentalimportance. For example, we know thatthe Japanese Fifth Generation programemphasizes the development of newmachines based on some principles ofterthan thc abstract model of Allan Turingand John Von Neumann.

All computcrs that exist today arebased on this abstract computationmodel. Unfortunately, there are noknown computational models which areshown to be more powerful than AllanTuring's abstract machine. On the otherhand, there is significant evidence thatnatural information processing systems,i.e. human and animal brains, aremassively parallel systems consisting ofsimple processing units. Research in thehuman thinking process will eventuallylead A.I. into the parallel computationdomain to a greater extent than today.

Regarding the innovative theoreticalmodels of the computer, we do not knowhow cost effective their implementationwill be. In any case, it is not somethingthat we will see in the near future.

New Centrefor Theoretical Research

in Bangalore

by K.S. Iayaraman

Courtesy ofNature, Vol 339, 22 Juru 1989

A Centre for advanced scientificresearch is to be set up in Bangalore andnamed after Jawaharial Nehru, India'sfirsr prime minister, whose centenary isbeing celebrated this year. The $16million centre will be structured alongthe lines of the international institute fortheorctical physics at Tricste in ltaly.

Profcssor CN.R. Rao, director of ttre"Indian Institute of Science (IIS), also inBangalore, is to act as president of theNehru Centre which, he says, "will bedevotcd to scientific research at thehighest level in frontier areas". It willset up its own facilities for theoreticalwork and will have access to theworkshop and laboratories of IIS forexperimental work.

The centre has been registered as anon-profit society with the primaryobjective of providing an excellentclimate for research and acting as anational and international forum for freeexchange of ideas through workshops,seminars and summer sQhools. It willhave a full-time faculty as well ashonorary and visiting positions. Raohopes the new centre will attract Indianintellectuals now drawn to Trieste.

Book review from IAE^A.

IAEA Yearbook 1989Publication date: August 1989. Price:Austrian Schillings 560,-. Code:STIiPUB/832.S u m m a r y : The Yearbook providesdescriptions of the IAEA's majorprogrammes, with articles on particularprojects and areas of activity, togetherwith reports of particular current interestand general information about the IAEA.The Yearbook presents the work of theIAEA in the context of scientific,technical and economic developmentsworldwide.

Nelvs from ICTP - No. 2'1125 - Jull'/August 1989

Contents: Foreword by the DirectorGeneral; the IAEA's Contribution toSustainable Development: Part A -Transfer of Nuclear Technology; Part B- Applications of Nuclear Techniques;Part C - Nuclear Power and Fuel Cycle:Status and Trends; Part D - NuclearSafety Review; Part E - IAEASafeguards; Part F - The IAEA.Parts A, B, C and D are also availableseparately.

IAEAWagramenhasse 5P.O. Box 1004-1400 ViennaAustria

Activities at ICTPFebruary - June 1989

Tit le : MINIWORKSHOP ONSTRONGLY C O R R E L A T E DELECTRON SYSTEMS INCONDENSED MATTER, 19 June - 21July.

Organizers: Professors G. Baskaran(Indian Institute of MathematicalSciences, Madras, India), A.E.Ruckenstein (University of California,La Jolla, USA), E. Tosatti (InternationalSchool for Advanced Studies, ISAS-SISSA, and ICTP, Trieste, Italy) and YuLu (P.R. ChinalCTP), with the co-sponsorship of the International Schoolfor Advanced Studies (ISAS-SISSA,Trieste, Italy).

Lectures: Introduction to metal-insulator transition and the strongcorrelation problem. Some results infermion mean field theories of the largeU Hubbard model. Itinerant vs. localizedAF of CuO2 layers within a slave bosonapproach. Gauge theory of dimer phasewith holes. Kondo effect: a saga ofstrongly interacting fermions. Instabilityof the ferromagnetic state in the large Ulimit of the Hubbard model. Novel localsymmetries and chiral symmetry brokenphases. Variational study of the pairhopping model. Exact diagonalizationstudies of small correlated systems. Twoquantum liquids. Quantum spin systemsand flux phases for pcdestrians.

The Miniworkshop was attendcd by67 lecturers apd participants (23 fromdeveloping countries).

Ti t te : SUMMER SCHOOL INHIGH ENERGY PHYSICS ANDCOSMOLOGY, 26 June - 18 August,

including the Workshop on supcrstrings(12 - 14 July) and the Workshop onphenomenology in high energy physicsand cosmoloSy (16 -18 August).

Organ ize r s : Professors G. Ellis(International School for AdvanccdStudies, ISAS-SISSA, Trieste, Italy),J.C. Pati (University of Maryland,College Park, USA), S. Randjbar-Daemi(Iran[CTP, Tricste, Italy), E. Sezgin(Turkey[CTP, Trieste, Italy) and Q.Shafi (University of Dclaware, Newark,USA), in collaboration with theInternational School for AdvancedStudies (ISAS-SISSA) and the ItalianInstitute for Nuclcar Physics (INFI'|.

Lectures: Part I: Introduction tostrings and conformal field theory.Multi-loop amplitudes in superstringtheory. Coulomb gas approach torational conformal field thcories. Stringtheory and 2-dim. quantum gravity.Landau-Ginzburg approach to N=2superconformal field theory. Conformalfield theory. N=2 string theory.

Part II: Physics of the early universe.Perturbative QCD. Probing the standardelectroweak model. Inflation. LatticeQCD. Superphysics. QCD withdynamical quarks. To grand unificationand beyond. Flavour physics.Superphysics. Scalar particls searchcs.Towards a unification of mass scalcs.Gauge theories and integrable models.Triviality and Higgs mass bound: astatus report. Inflation and quantumcosmology. Interacting higher-spinactions from strings. Cosmologicalconstant. Highlights of GRl2. Thesurface tension in hadron nucleation andother first ordcr phase transitions. Stellarstructure, evolution and the solarneutrino problem (SNP). MicrowaveBackground radiation and large scalestructure of the universe. WIMPs and theSNP. Current slatus of the strong CPproblem. Three generation superstringmodels. Conformal field theory in theupper half plane. WIMPs and the solarpulsation problem (SPP). WIMPs inother slars: predictions and observations.4-dim. superstring models. Applicationof conformal field theory to string fieldtheory.

Workshop on superstrings:Fundamental length scale in stringtheory. The quartic effective action of theheterotic string and supersymmetry.Heterotic superstrings vacua based onnon-compact currcnt algcbras. Topologyof singularities and phase diagram onN=2 superconformal field theories.Contour integral representation of

characters and supercharacters in rationalconformal field ttreories. Real weight B-C systems and conformal field theory inhigher genus. Massless and massivespinning superparticles. Duality andquantum groups. Minimal model inhigher genus. Fusion and braiding in W-algebra extended and coset conformaltheories. Deforming the Venezianomodel ("q-strings"). The role of 2-dim.quantum gravity in string theory.General SU(2) Coset models andproducts of Virasoro models. BRS -invariance in unusual settings.Anomalies of the heterotic sigma modclsin a curved (1,0) superspace. Topologicalaspects of CalabiYau compactifications.Landau-Ginzburg Hamiltonians,expansions and N=2 superconformalmodcls. Hidden symmetries in stringtheories. Non-perturbative aspects andexact results for N=2 Landau-Ginzburgtheories. Integrating out Kaluza-Kleinmodes in string compactifications.Quantization of dynamical systems incurved phase space. Quantization ofgeneral gauge theories in curved phasespace. Uq[sl(2)] invariant operators andminimal theories fusion matrices.

Workshop on phenomenology inhigh energy physics and cosmology:Experimental status of heavy flavourphysics. Proton spin. The status on CPviolation. Physics from the Mark IIDetector and the SLC. Overview of ftreegeneration Calabi-Yau models. Strings athigh temperature. The cosmologicalconsuant with or without wormholes. Anatural connection between the origin ofinflation and the hierarchy in mass-scales. Physics at LEP. 4-dim.suporstring models. Physics at theFermilab Tevatron collider.

The School was attended by 239lecturcrs and participants (162 fromdeveloping countdes).

Tit le: QUASICRYSTALS, 4 - 7July.

Organizers: Profcssors M.V. Jari0(Texas A. & M. University, CollegeStation, USA) and S.O. Lundqvist(Chalmers University of Technology,GOtcborg, Sweden), with the co-sponsorship of the International Schoolfor Advanced Studies (ISAS-SISSA,Trieste, Italy).

Lectures: Recent developments inthe study of quasicrystals. Superlattice,microquasicrystals and approximants.Quasicrysul structure determination -Al-Cu-Li. The structure of quasicrystals:

News from ICTP - No. 21125 - July/August 1989

advantages and limits of a neutrondiffraction approach. Icosahedral alloyswithout phason disorder. Micro-crystalline state of pseudo-icosahedralsymmetry, crystalline to quasicrystallinephase Eansformation in the Al-Cu-Li andAl-Cu-Li and Al-Cu-Fe systems.Octagonal and dodecagonal quasicrystals.Quasiperiodic arrays in thin films of Al-Mn. On the symmetry of tilings andquasicrystals. Characteizations of orderin aperiodic structures and arithmeticalproperties. Entropically stabilizedquasicrystals. Tetracoordinatedquasicrystals. A new algorithm for thegeneration of quasicrystals.Quasicrystals, approximations anddynamical systems. Simplequasicrystalline tilings. Quasicrystalsversus modulated crystals. Topology ofatomic surfaces in icosahedralquasicrystals. Modelling quasicrystalgrowth. Phason elasticity and surfaceroughening. Quasicrystal grainboundaries from ttre entropic and density-wave viewpoints. Soft modecrystallization. Why only quadraticirrationals are observed in quasicrystals?Kinetic roughening of quasicrystals.Defective vertex configurations inquasicrystals. Displacive mappings fromquasiperiodic to periodic patterns.Dislocations in decagonal Al-Cu-Coalloy. Multifractal analysis oflocalization problem in quasiperiodicsystems. Real space renormalization andcritical properties of Fibonaccipotentials. Electronic structure ofquasicrystals. On the electronicproperties of a three-dimensionalquasicrystal with weak potential.Diamagnetism in quasicrystallinesuperconductors. Algebraic symmetriesin quasicrystal scattering data. Crossoverbehaviour from 'extended' to chaoticwavefunctions in 1d quasicrystals.Quasiperiodic Hume-Rothery stablealloys in 1d: an exact solution. Whereand how large are the gaps in spectra ofdeterministic aperiodic systems? Noveltransition between critical and localizedstates in a ld quasiperiodic system.Thermal properties of a classical modelof a 2d quasicrystal. Vibrationalproperties of Fibonacci chains withoutand with defects.

The Conference was attended by 62lecturers and participants (12 fromdeveloping countries).

Title: TRIESTE CONFERENCEON SUPERMEMBRANES AND

PHYSICS IN 2+1 DIMENSIONS, 17 -21 July 1989.

Organizers: Professors M. Duff(Texas A&M University, USA), C.Pope (Texas A&M University, USA)and E. Sczgin (Turkey[CTP, Trieste,Italy).

Lectures: Membranes, extendedobjects and infinite-component waveequations. Supermembranes: a review.Towards a conformal field theory forhigher spins. Infinite dimensionalmembrane symmetries ttrat generalize theconformal symmetry of the string.Evolution equation of strings andmembranes. Singleton and doubletonsupermultiplets of spacetimesupergroups and infinite spinsuperalgebras. Worldvolumesupers ymmetry from spacetimesupersymmetry. Braiding matrices,modular transformations and topologicalfield theory in 2+1 dimensions. Newsymmetries in 2+l dimensions.Quantum topologically massive gravity.Topology and 3-dimensional fieldtheories. Area-preserving algebras andmembrane topology. Geometry andtopology of 3-manifolds. Non-AbelianChern-Simons theory, conformal fieldrheory and knot polynomials. Covariantcanonical quantization of superparticles.The light cone gauge. 2+1 topologicalterms, anyons and their possible physicalapplication in high Ts superconductivity.The supermembrane mass spectrum.New directions in supermembranes.Strings from membranes: the origin ofconformal invariance. New spacetimesuperalgebras and their Kac-Moodyextension. Consistent dynamics ofhigher spin gauge fields. Comments onintegrability in 2+1. The large N limitof extended conformal symmetries. Area-preserving diffeomorphisms andsupermembrane Lorentz invariance. TheE9 current of d=2, N=16 supergravity.Super Yang-Mills quantum mechanicsand supermembrane spectrum.Trigonometrical structure constant. fortle classical Lie algebras and the vacuumstates SU(-) Yang-Mills theory. On thecovariant quantization of gauge theoriesincluding supermembranes. BRSTquantization of the Green-Schwarzsuperstring. Chern-Simons supergravity.Chern-Simon theories of symplecticsuper-diffeomorphisms. Structure ofSDiff, the weather forecast and solaractivity. Virasoro algebra and amembrane symmetry. Gauge theory ofsingletons: results and perspectives. Tealtunnelling geometries and ttre large scale

topology of the universe. Generalizedsigma models and the index theorem.Weyl-invariance of membranes.Harmonic superspace in supersymmetryand beyond. Supermembranes on blackholes. Generic curved spacesupermembrane theories. Concludingremarks.

The Conference was attended by 35lecturers and participants (6 fromdeveloping countries).

T i t l e : S T R O N G L YCORRELATED ELECTRONSYSTEMS, 18 - 21 July.

O r g a n i z e r s : Professors A.Ruckenstein (University of California,San Diego-La Jolla, USA), E. Tosatti(International School for AdvancedStudies, ISAS-SISSA, Trieste, Italy) andYu Lu (ICTP), with the co-sponsorshipof the International School for AdvancedStudies (ISAS-SISSA, Triesre, Iraly).

Lectures: Phenomeno logicalconstraints on theories of high Tgsupcrconductors. Strong correlation andnovel superconducting states. Theory ofmetal-insulator transition. The spiralphase and superconductivity in the dopedHeisenberg model. Quasiparticles indoped planar antiferromagnet. Self-consistent renormalized motion of holesin a quantum antiferromagnet. Numericalstudies of the t-J model: exact groundstate and flux phases. Finite sizecalculations on the 2D Hubbard model.Fermion quantum Monte Carlo andHubbard model. Preliminary grouptheory analysis of a 4x4 two-dimensionalHubbard model. Properties of two-dimensional systems with fractionalstatistics. Anyons on a lattice.Superfluidity of the anyon gas. Spinsinglet fractional quantized Hall states. Amodel for orbital antiferomagnetism.TwisLed superconductors and twistcdmagnets. Magnetism (+ super-conductivity?) in the 2D Hubbard model.Intersite conelation in variational wavefunctions. New exactly tractable limit forcorrelated lattice electrons. Conelatedfermions on a Iattice in high dimensions.Heavy fermions: past, present and future.Understanding of NMR in high Tgmaterials. Slave boson approach to theHubbard model. Itinerant vs. localizedAF of CuO2 layers within a slave bosonapproach. Instability of the ferromagneticscate in the large U limit of the Hubbardmodel. Strongly correlated, quasi-one-dimensional bands: group states, optics,and phonons. Nematic and dimer states

Nervs from ICTP - No. 21125 - July/August 1989

in low dimensional magnets. Ramanspectra and strong O-O fluctuations inYBa2Cu3O7. Topological statisticstransmutation and superconductivity.Treatment of strong correlations byprojection techniques. Pairing theory intwo-dimensions. Importance of spin-flipprocesses in quantum antiferromagnets.One loop corrections to the Gutzwillerapproximation (within the slave bosonapproach). Quasiparticles andphotoemission spectra in correlatedelectron systems. Binding of holons inthe chiral spin states. Thesupersymmetric Hubbard model and highTg superconductivity. The effect oftopological excitations in the two-dimensional quantum Heisenbergantiferromagnet. 4KF correlations in lDsystems revisited. U = - Hubbard model:towards the exact solution in d = - .

Thc Course was attended by 79lecturers and participants (22 fromdeveloping countries).

T i t l e : I N T E R N A T I O N A L

SYMPOSruM ON HIGHLICHTS INCONDENSED MATTER PHYSICS(dedicated to Prof. K.S. Singwi on theoccasion of his 70th birthday inrecognition of his extensive andpioneering work in the field), 1 - 3August.

Organizers : Professors P. Yena(Virginia Commonwealth University,Richmond, USA), R. Kalia (ArgonneNational Laboratory, Argonne, USA),M.P. Tosi (University of Trieste andICTP, Italy) and P. Vashishta (ArgonneNational Laboratory, Argonne, USA).

Lec tu re s : Local field effects inweakly to strongly interacting Fermisystems. Applicability of the local fieldconcept for the elcctron gas. Dynamicsusceptibility of a free electron gas in Ddimensions and its analytic properties.Hubbard corrections in a 2-D electrongas. A novel approach to correlationeffects in electron spectroscopies. Ondynamic conelations in uniform electronliquid. The puzzling neutrinos.Dynamics of classical and quantumliquids from neutron scatteringexperiments. Relaxations in supercooled

Lennard-Jones systems. Dynamicalcorrelation function and transportcoefficients of dense fluids. The structureand dynamics of intercalated brnarysystems. Interatomic potential energy.Electron correlations and dielectricscreening in a layered electron gas.Impurity problem in metals. Polaron gasin low dimensional structures. A coldstart to the Universe? Density functionaltheory of superconductors. Ground-staleenergy of the one-and two-componentHubbard model calculated by the methodof Singwi, Tosi, Land and Sjolander.Superconducting correlations in tle newhigh Tg superconductors from a study ofdynamic response function. Hydrogen inmetals. Density wave tleory of Wignercrystallization. Electron-hole liquids andmany-body theories.

The Symposium was attended by 26lecturers and participants (5 fromdeveloping countries).

Future Activities at ICTP in 1989

Computations in Physics and and Physics in Computation(Anniversarv Adriatico Research Conference)Adriatico Working Party on Condensed Matter Properties of Neutron StarsWorkshop on Materials Science and Physics of Nonconventional Energy SourcesConference on Lasers in ChemistrvWorkshop on Interaction between Physics and Architecturein Environment Conscious DesienTrieste Conference on Recent DeveloDments in Conformal Field TheoriesFifth Collese on Microprocessors: Technology and Arrrrlications in PhysicsWorkshop on Soil PhysicsCollese on Differential Geomelry25th Anniversary Conference on "Frontiers in Physics, High Technologyand Mafrematics"Workshoo on TelematicsICTP & INFN Course on Basic VLSI Desicn TechniquesThird Autumn Workshop on "Atmospheric Radiation and Cloud Physics"

5 - 8 Septemberl l - 29 Septemberl l - 29 Seplember1 8 - 2 2 Septernber

25 - 29 Serrtember2 - 4 October

2 - 27 October9 - 27 October

30October- I December

3l October - 3 Novembcr6 - 24 November

6 November - 1 December27 November -15 December

For information and applications to courses, kindly write to the Scientific Programme Office.

Intemational Centre for Theoretical Physicsof IAEA and UNESCOStrada Costiera, l1P.O. Box 586l-34136 Trieste, Italy

Telephone: $ q 22.401Cable: CENTR ATOM

Telex: 46A392 ICTP ITelefax: (40) 22.41.63

Bitnen SYSTEM@ITSICTP.BITMT

EDITORIAL NOTE - News from ICTP is not an official document of the International Centre for Theoretical Physics. Its purpose is to keepscientists informed on past and future activities at the Centre and initiatives in their home countries. Suggestions and criticisms shouldbe addressed to Dr. A.M. Hamende, Scientific Information Officer.