BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Structural\, Temporal and Semantic Information - Prof. Wojciech Sz
 pankowski\, Purdue University
DTSTART:20241016T130000Z
DTEND:20241016T140000Z
UID:TALK218494@talks.cam.ac.uk
CONTACT:Prof. Ramji Venkataramanan
DESCRIPTION:Shannon's information theory has served as a bedrock for advan
 ces in communication and storage systems over the past five decades. Howev
 er\, this theory does not handle well higher order structures (e.g.\, grap
 hs\, geometric structures)\, temporal aspects (e.g.\, real-time considerat
 ions)\, and semantics (quoting Shannon ``the semantic aspects of communica
 tion are irrelevant to the engineering problem '') While accepting Shannon
 's assertion\, we observe that since Shannon's days information has emerge
 d as the currency of our modern technological society.\n\nIn this talk\, w
 e present some recent results on structural temporal information and seman
 tic information. We first show how to extract temporal information in dyna
 mic networks (arrival of nodes) from its structure (unlabeled graphs). We 
 then proceed to establish fundamental limits on information content for so
 me data structures\, and present asymptotically optimal lossless compressi
 on algorithms achieving these limits for various graph models. Finally\, w
 e deal with semantic and study a simple but profoundly important setting i
 n which a sender wants to relay to a recipient a minimal summary of its da
 tabase of knowledge\, written in the language of propositional logic\, so 
 that on the receiving side one can answer the same queries as the sender.\
 n\n--------------------- bio sketch ---------------------\n\nWojciech Szpa
 nkowski is Saul Rosen Distinguished Professor of Computer\nScience at Purd
 ue University where he teaches and conducts research in\nanalysis of algor
 ithms\, information theory\, analytic combinatorics\,\nnetwork science\, r
 andom structures\, and stability problems of\ndistributed systems.  He hel
 d several Visiting Professor/Scholar\npositions\, including McGill Univers
 ity\, INRIA\, France\, Stanford\,\nHewlett-Packard Labs\, Universite de Ve
 rsailles\, University of\nCanterbury\, New Zealand\, Ecole Polytechnique\,
  France\, the Newton\nInstitute\, Cambridge\, UK\, ETH\, Zurich\, and Jagi
 ellonian University.\nKrakow\, Poland.  He has been on editorial boards on
  many journals\nincluding IEEE Trans. Information Theory\, TCS\, CPC\, and
  ACM Trans. on\nAlgorithms.  Szpankowski is a Fellow of IEEE\, and the Ers
 kine Fellow.\nIn 2010 he received the Humboldt Research Award\, in 2015 he
  was\nrecipient of the naugural Arden L. Bement Jr. Award\, and in 2020 he
 \ndelived the Philippe Flajolet Lecture.  He published two books:\n"Averag
 e Case Analysis of Algorithms on Sequences"\, John Wiley & Sons\,\n2001\, 
 "Analytic Pattern Matching: From DNA to Twitter" (with P.Jacquet)\n\, Camb
 ridge\, 2015\, and ``Analytic Information Theory: From Compression\nto Lea
 rning'' (with (M. Drmota)\, 2023.  In 2008 he launched the\ninterdisciplin
 ary Institute for Science of Information\, and in 2010 he\nbecame the Dire
 ctor of the NSF Science and Technology Center for Science of Information.
LOCATION:MR5\, CMS Pavilion A
END:VEVENT
END:VCALENDAR