BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:Asia/Kolkata BEGIN:STANDARD DTSTART:19451014T230000 TZOFFSETFROM:+0630 TZOFFSETTO:+0530 TZNAME:IST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20200104T104242Z UID:F7DFBDF2-35FC-471F-9458-1380833AA1B7 DTSTART;TZID=Asia/Kolkata:20181205T090000 DTEND;TZID=Asia/Kolkata:20181207T170000 DESCRIPTION:The main idea of this school is to raise awareness and applicab ility about Deep Learning (DL)\nwith Computational Intelligence (CI) among the research communities. Traditional learning\nand computing methods dea l with vast range of applications related to reasoning\, decision\nmaking\ , perception building etc. However\, DL with CI deals with dynamical syste ms more\nefficiently by embedding and facilitating learning mechanism. Mos t of the data obtained in real\ntime environment from various domains i.e. environment\, industry\, business\, biology involves\nlots of imprecision and vagueness. To address these issues in various domains of algorithms\, \ntools or techniques are required which should be adaptive and robust so that they can handle\nthe uncertainty and dynamic nature of this system in efficient and optimized way.\n“CI is a set of biological and linguistic tools and methodologies to address complex real-world\nproblems to which traditional Artificial Intelligence (AI) approaches may not be very effect ive.\nCI comprises of concepts and implementations that ensures intelligen t behaviour in complex\nand dynamic environment.”\n\nAccording to Robert J. Marks\, “Neural Network\, Genetic Algorithms\, Fuzzy systems\,\nevol utionary programming and artificial life are the building blocks of Comput ational\nIntelligence.”\nUsing CI tools\, we will be able to build the s ystems which are prone to adaption\, robust across\nproblem domains\, appl y extrapolated reasoning and behave intelligently in given state. Neural\n network techniques provide capability of computational adaption. System ca n improve its\nparameter without any intervention based on optimizing crit eria same as human learning\noccurs. Fuzzy systems help in defining the sy stem where we have a rough estimate of system\nrequirements. Evolutionary algorithms are good enough to optimize parameters and to select\nbest amon g given constraints. Synergistic effect of these tools may increase their individual\nperformances and gives better adaptive and reliable system. Kn owledge representation\,\nreasoning\, information mining\, discovery scien ce\, web intelligence\, semantic web\, multi agent\nsystems and designing of products i.e. air conditioners\, automobile systems\, ABS\, cameras\,\n dishwashers\, pattern recognition in remote sensing\, video games are the major areas where CI\ncan be very helpful.\nAdvantages of DL and CI over e xisting deep learning algorithms: One has automatic structure\noptimizatio n ability where a neural network structure (e.g.\, the number of layers\, the number\nof units in each layer\, the type of an activation function at each unit\, etc.) is automatically\noptimized for a given data set and a given objective by an evolutionary structure learning\ntechnique. The othe r is multi-objective ability where several different neural networks are\n simultaneously obtained under a multi-objective scenario. Various multi-ob jective\nformulations can be considered for deep learning. A general formu lation is a combination of\ncomplexity minimization and accuracy maximizat ion. For detection problems\, false positive\nand false negative can be ha ndled as separate objectives.\nThe school will bring people working in CI and DL domain to a common platform for\ngenerating innovative ideas. The s chool will also assess the state of the art on what new\ndirections lie op en for research in area of CI and DL. In this way\, the school will genera te\nexciting new communication across various CI and machine learning disc iplines.\nThe school has attracted around 99 participants (Annexure I) fro m various engineering\ncolleges\, industries and organizations across the India. In nutshell\, the event was an excellent\nopportunity where thought -provoking lectures were conducted for fruitful interaction and\nseveral t echnical challenges. It helped exciting new communication across various D L and CI\ndisciplines and helps to define an emerging international resear ch community.\n\nKanpur\, Uttar Pradesh\, India\, 208016 LOCATION:Kanpur\, Uttar Pradesh\, India\, 208016 ORGANIZER:nishchal@iitk.ac.in SEQUENCE:2 SUMMARY:2018 IEEE CIS Summer School on Deep Learning and Computational Inte lligence URL;VALUE=URI:https://events.vtools.ieee.org/m/217259 X-ALT-DESC:Description: <br /><p>The main idea of this school is to raise a wareness and applicability about Deep Learning (DL)<br />with Computationa l Intelligence (CI) among the research communities. Traditional learning<b r />and computing methods deal with vast range of applications related to reasoning\, decision<br />making\, perception building etc. However\, DL w ith CI deals with dynamical systems more<br />efficiently by embedding and facilitating learning mechanism. Most of the data obtained in real<br />t ime environment from various domains i.e. environment\, industry\, busines s\, biology involves<br />lots of imprecision and vagueness. To address th ese issues in various domains of algorithms\,<br />tools or techniques are required which should be adaptive and robust so that they can handle<br / >the uncertainty and dynamic nature of this system in efficient and optimi zed way.<br />&ldquo\;CI is a set of biological and linguistic tools and m ethodologies to address complex real-world<br />problems to which traditio nal Artificial Intelligence (AI) approaches may not be very effective.<br />CI comprises of concepts and implementations that ensures intelligent be haviour in complex<br />and dynamic environment.&rdquo\;</p>\n<p>According to Robert J. Marks\, &ldquo\;Neural Network\, Genetic Algorithms\, Fuzzy systems\,<br />evolutionary programming and artificial life are the buildi ng blocks of Computational<br />Intelligence.&rdquo\;<br />Using CI tools\ , we will be able to build the systems which are prone to adaption\, robus t across<br />problem domains\, apply extrapolated reasoning and behave in telligently in given state. Neural<br />network techniques provide capabil ity of computational adaption. System can improve its<br />parameter witho ut any intervention based on optimizing criteria same as human learning<br />occurs. Fuzzy systems help in defining the system where we have a rough estimate of system<br />requirements. Evolutionary algorithms are good en ough to optimize parameters and to select<br />best among given constraint s. Synergistic effect of these tools may increase their individual<br />pe rformances and gives better adaptive and reliable system. Knowledge repres entation\,<br />reasoning\, information mining\, discovery science\, web i ntelligence\, semantic web\, multi agent<br />systems and designing of pro ducts i.e. air conditioners\, automobile systems\, ABS\, cameras\,<br />di shwashers\, pattern recognition in remote sensing\, video games are the ma jor areas where CI<br />can be very helpful.<br />Advantages of DL and CI over existing deep learning algorithms: One has automatic structure<br />o ptimization ability where a neural network structure (e.g.\, the number of layers\, the number<br />of units in each layer\, the type of an activati on function at each unit\, etc.) is automatically<br />optimized for a giv en data set and a given objective by an evolutionary structure learning<br />technique. The other is multi-objective ability where several different neural networks are<br />simultaneously obtained under a multi-objective scenario. Various multi-objective<br />formulations can be considered for deep learning. A general formulation is a combination of<br />complexity m inimization and accuracy maximization. For detection problems\, false posi tive<br />and false negative can be handled as separate objectives.<br />T he school will bring people working in CI and DL domain to a common platfo rm for<br />generating innovative ideas. The school will also assess the s tate of the art on what new<br />directions lie open for research in area of CI and DL. In this way\, the school will generate<br />exciting new com munication across various CI and machine learning disciplines.<br />The sc hool has attracted around 99 participants (Annexure I) from various engine ering<br />colleges\, industries and organizations across the India. In nu tshell\, the event was an excellent<br />opportunity where thought-provoki ng lectures were conducted for fruitful interaction and<br />several techn ical challenges. It helped exciting new communication across various DL an d CI<br />disciplines and helps to define an emerging international resear ch community.</p> END:VEVENT END:VCALENDAR