BEGIN:VCALENDAR
VERSION:2.0
PRODID:-// - ECPv6.16.2//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://art-ai.dmu.ac.uk
X-WR-CALDESC:Events for 
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:20170326T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:20171029T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:20180325T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:20181028T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:20190331T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:20191027T010000
END:STANDARD
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:20200329T010000
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:20201025T010000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T183000
DTEND;TZID=Europe/London:20190523T193000
DTSTAMP:20260520T033805
CREATED:20190426T162136Z
LAST-MODIFIED:20190430T164018Z
UID:2077-1558636200-1558639800@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : Phoenix VR - Rose Coloured
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/rose-coloured-2/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Rose-Colored-600.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T173000
DTEND;TZID=Europe/London:20190523T183000
DTSTAMP:20260520T033805
CREATED:20190426T151854Z
LAST-MODIFIED:20190429T154330Z
UID:2050-1558632600-1558636200@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : Phoenix VR - Rose Coloured
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/rose-coloured/
LOCATION:Leicestershire
CATEGORIES:Conference,Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Rose-Colored-600.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T153000
DTEND;TZID=Europe/London:20190523T161500
DTSTAMP:20260520T033805
CREATED:20190426T151148Z
LAST-MODIFIED:20190501T073129Z
UID:2048-1558625400-1558628100@art-ai.dmu.ac.uk
SUMMARY:
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/panel-state-of-the-ai-art/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/state-of-the-ai-art.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T144500
DTEND;TZID=Europe/London:20190523T153000
DTSTAMP:20260520T033805
CREATED:20190426T123435Z
LAST-MODIFIED:20190516T202809Z
UID:1994-1558622700-1558625400@art-ai.dmu.ac.uk
SUMMARY:Keynote : Jake Elwes - Creative Dilemmas
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/keynote-jake-elwes-creative-dilemmas/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Jake-Elwes.jpeg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T140000
DTEND;TZID=Europe/London:20190523T143000
DTSTAMP:20260520T033805
CREATED:20190426T124710Z
LAST-MODIFIED:20190430T143635Z
UID:1998-1558620000-1558621800@art-ai.dmu.ac.uk
SUMMARY:Talk : State of the AI-Art
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/talk-state-of-the-ai-art/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/elliot_turing_test_square.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T130000
DTEND;TZID=Europe/London:20190523T133000
DTSTAMP:20260520T033805
CREATED:20190426T131714Z
LAST-MODIFIED:20190430T002302Z
UID:2005-1558616400-1558618200@art-ai.dmu.ac.uk
SUMMARY:Talk : Love Apparatus
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/love-apparatus-2019/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Fabrizio-Augusto-Poltronieri-5859.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T130000
DTEND;TZID=Europe/London:20190523T133000
DTSTAMP:20260520T033805
CREATED:20190426T130756Z
LAST-MODIFIED:20190430T214731Z
UID:2002-1558616400-1558618200@art-ai.dmu.ac.uk
SUMMARY:Talk : Making a Chatbot
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/making-a-chatbot/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/PWMay17CAR-heaney_600-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T111500
DTEND;TZID=Europe/London:20190523T120000
DTSTAMP:20260520T033805
CREATED:20190426T114758Z
LAST-MODIFIED:20190430T002522Z
UID:1985-1558610100-1558612800@art-ai.dmu.ac.uk
SUMMARY:Keynote : Mario Klingemann - Interstitial Space 2019
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/keynote-mario-klingemann-interstitial-space-2019/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/mario-klingemann-570x570.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190523T104500
DTEND;TZID=Europe/London:20190523T231500
DTSTAMP:20260520T033805
CREATED:20190426T094116Z
LAST-MODIFIED:20190430T002551Z
UID:1926-1558608300-1558653300@art-ai.dmu.ac.uk
SUMMARY:Keynote : Ernest Edmonds - Art and Influence: Learning in Augmented Worlds
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/art-and-influence-learning-in-augmented-worlds/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Conference,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Ernest-pic-square.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190518T193000
DTEND;TZID=Europe/London:20190518T233000
DTSTAMP:20260520T033805
CREATED:20190426T160035Z
LAST-MODIFIED:20190429T234803Z
UID:2059-1558207800-1558222200@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : Terminator 1 & 2
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/terminator12/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/arnie600.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190517T183000
DTEND;TZID=Europe/London:20190517T194500
DTSTAMP:20260520T033805
CREATED:20190426T100347Z
LAST-MODIFIED:20190430T160258Z
UID:1929-1558117800-1558122300@art-ai.dmu.ac.uk
SUMMARY:Performance : Human+Machine - Improbotics
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/performance-improbotics/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Launch Event,Performance
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/improbotics_humans_square.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190517T173000
DTEND;TZID=Europe/London:20190517T183000
DTSTAMP:20260520T033805
CREATED:20190426T113312Z
LAST-MODIFIED:20190430T154555Z
UID:1982-1558114200-1558117800@art-ai.dmu.ac.uk
SUMMARY:VIP Launch : Gallery
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/vip-launch-gallery/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Installation,Launch Event
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Britbot-and-Interstitial-Space.png
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190517T160000
DTEND;TZID=Europe/London:20190517T170000
DTSTAMP:20260520T033805
CREATED:20190426T092918Z
LAST-MODIFIED:20190429T234501Z
UID:1920-1558108800-1558112400@art-ai.dmu.ac.uk
SUMMARY:Lecture : The Human in the Design
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/the-human-in-the-design/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Launch Event,Talk
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Gerhard-fischer-professor-photo.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190516T093000
DTEND;TZID=Europe/London:20190531T200000
DTSTAMP:20260520T033805
CREATED:20190425T224227Z
LAST-MODIFIED:20190501T075750Z
UID:1870-1557999000-1559332800@art-ai.dmu.ac.uk
SUMMARY:Installation : Britbot
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/britbot-by-libby-heaney/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Installation
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Black-Britbot.png
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20190516
DTEND;VALUE=DATE:20190601
DTSTAMP:20260520T033805
CREATED:20190430T010553Z
LAST-MODIFIED:20190430T010554Z
UID:3042-1557964800-1559347199@art-ai.dmu.ac.uk
SUMMARY:Installation : Neural Synthesis
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/neural-synthesis/
LOCATION:Leicester Central Library\, Bishop Street\, Leicester\, Leicestershire\, LE1 6AA
CATEGORIES:Installation
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Kogan-Colours-1.png
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20190516
DTEND;VALUE=DATE:20190601
DTSTAMP:20260520T033805
CREATED:20190430T005534Z
LAST-MODIFIED:20190430T005639Z
UID:3037-1557964800-1559347199@art-ai.dmu.ac.uk
SUMMARY:Installation : Working with Useless Machines
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/useless-machines/
LOCATION:Haymarket Theatre\, 1 Garrick Walk\, Leicester\, LE1 3AF\, United Kingdom
CATEGORIES:Installation
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Nadine-Lessio-Smoothie.png
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20190516
DTEND;VALUE=DATE:20190601
DTSTAMP:20260520T033805
CREATED:20190430T004742Z
LAST-MODIFIED:20190501T080847Z
UID:3031-1557964800-1559347199@art-ai.dmu.ac.uk
SUMMARY:Installation : Neural Zoo
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/neural-zoo/
LOCATION:Highcross Shopping Centre\, 5 Shires Lane\, Leicester\, LE1 4AN\, United Kingdom
CATEGORIES:Installation
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Crespo-Green-1.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20190516
DTEND;VALUE=DATE:20190601
DTSTAMP:20260520T033805
CREATED:20190430T004049Z
LAST-MODIFIED:20190501T080223Z
UID:3026-1557964800-1559347199@art-ai.dmu.ac.uk
SUMMARY:Installation : Neural Zoo
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/neural-zoo-2/
LOCATION:Haymarket Shopping Centre\, 1 Kildare Street\, Leicester\, LE1 3YA
CATEGORIES:Installation
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/the_box_3467-e1556640646399.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;VALUE=DATE:20190516
DTEND;VALUE=DATE:20190601
DTSTAMP:20260520T033805
CREATED:20190425T175819Z
LAST-MODIFIED:20190430T003047Z
UID:1853-1557964800-1559347199@art-ai.dmu.ac.uk
SUMMARY:Installation : Interstitial Space 2019
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/interstitial-space-2019/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Installation
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/04925-1-e1556297323920.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190512T105000
DTEND;TZID=Europe/London:20190512T122000
DTSTAMP:20260520T033805
CREATED:20190426T161512Z
LAST-MODIFIED:20190429T225028Z
UID:2074-1557658200-1557663600@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : WALL-E
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/wall-e-3/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Wall-E-600-square.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190511T114000
DTEND;TZID=Europe/London:20190511T131000
DTSTAMP:20260520T033805
CREATED:20190426T161358Z
LAST-MODIFIED:20190429T224914Z
UID:2071-1557574800-1557580200@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : WALL-E
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/wall-e-2/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Wall-E-600-square.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20190511T110000
DTEND;TZID=Europe/London:20190511T123000
DTSTAMP:20260520T033805
CREATED:20190426T160942Z
LAST-MODIFIED:20190429T162053Z
UID:2063-1557572400-1557577800@art-ai.dmu.ac.uk
SUMMARY:Flim Screening : WALL-E
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/wall-e/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2019/04/Wall-E-600-square.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180513T111500
DTEND;TZID=Europe/London:20180513T124500
DTSTAMP:20260520T033805
CREATED:20180411T214303Z
LAST-MODIFIED:20190425T212420Z
UID:1410-1526210100-1526215500@art-ai.dmu.ac.uk
SUMMARY:War Games
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/war-games-2/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/WarGames-3.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180512T160000
DTEND;TZID=Europe/London:20180512T180000
DTSTAMP:20260520T033805
CREATED:20180405T144817Z
LAST-MODIFIED:20180412T235130Z
UID:1061-1526140800-1526148000@art-ai.dmu.ac.uk
SUMMARY:Ghost in the Shell
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/ghost-in-the-shell-2/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/Ghost-in-the-Shell-Anime.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180512T120000
DTEND;TZID=Europe/London:20180512T150000
DTSTAMP:20260520T033805
CREATED:20180401T214354Z
LAST-MODIFIED:20180406T020236Z
UID:186-1526126400-1526137200@art-ai.dmu.ac.uk
SUMMARY:Robot Maze
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/robot-maze/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Workshop
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/hex_bugs_1000px-1000x650.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180512T110000
DTEND;TZID=Europe/London:20180512T130000
DTSTAMP:20260520T033805
CREATED:20180406T012306Z
LAST-MODIFIED:20180412T235434Z
UID:1217-1526122800-1526130000@art-ai.dmu.ac.uk
SUMMARY:War Games
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/war-games/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/WarGames-3.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180511T203000
DTEND;TZID=Europe/London:20180511T223000
DTSTAMP:20260520T033805
CREATED:20180406T012705Z
LAST-MODIFIED:20180412T234804Z
UID:1223-1526070600-1526077800@art-ai.dmu.ac.uk
SUMMARY:Ghost in the Shell
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/ghost-in-the-shell/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/Ghost-in-the-Shell-Anime.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180510T183000
DTEND;TZID=Europe/London:20180510T203000
DTSTAMP:20260520T033805
CREATED:20180401T214209Z
LAST-MODIFIED:20180410T182927Z
UID:184-1525977000-1525984200@art-ai.dmu.ac.uk
SUMMARY:Curating the Machine
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/curating-the-machine/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Talk
ATTACH;FMTTYPE=image/gif:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/curating-the-machine-e1522922110125.gif
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180509T200000
DTEND;TZID=Europe/London:20180509T220000
DTSTAMP:20260520T033805
CREATED:20180402T090902Z
LAST-MODIFIED:20180412T235001Z
UID:225-1525896000-1525903200@art-ai.dmu.ac.uk
SUMMARY:Alpha Go
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/alpha-go/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Film Screening
ATTACH;FMTTYPE=image/jpeg:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/AG_Horizontal_Primary.width-440.jpg
END:VEVENT
BEGIN:VEVENT
DTSTART;TZID=Europe/London:20180509T183000
DTEND;TZID=Europe/London:20180509T200000
DTSTAMP:20260520T033805
CREATED:20180401T213614Z
LAST-MODIFIED:20180410T182824Z
UID:183-1525890600-1525896000@art-ai.dmu.ac.uk
SUMMARY:#Love Apparatus Talk
DESCRIPTION:Gene Kogan's Neural Synthesis is a video artwork that explores and expands upon the technique popularly known as ‘deep dream’. This is an iterative process for optimising the pixels of an image to obtain a desired activation state in a trained convolutional network. It primarily experiments with the dynamics of feedback-generated ‘deep dream’ videos\, where each frame in the film is initialised by the previous one. The novel aesthetics are achieved by ‘gating’ (or ‘masking’) the pixel gradients of multiple channels and mixing them using predetermined masking patterns\, while simultaneously distorting   the input canvas. \nA number of strategies used to create the artwork are directly inspired by the initial work of Google’s Deep Dream implementation\, particularly the work of Mike Tyka who first experimented with feedback\, canvas distortion (zooming) and mixing gradients of multiple channels. \nThe trained network used is Google’s Inceptionism network. The workflow for generating the artwork is under continuous development: future improvements include a more generalised canvas distortion function and improved masking from source images. \nAs part of the ART AI Festival 2019 artwork trail\, Gene Kogan's Neural Synthesis is on display at Leicester Central Library\, Bishop Street\, Leicester\, LE1 6AA - download map.
URL:https://art-ai.dmu.ac.uk/event/love-apparatus-talk/
LOCATION:Phoenix\, Phoenix Arts Centre\, 4 Midland Street\, Leicester\, LE1 1TG\, United Kingdom
CATEGORIES:Talk
ATTACH;FMTTYPE=image/png:https://art-ai.dmu.ac.uk/wp-content/uploads/2018/04/love_apparatus_twitter.png
END:VEVENT
END:VCALENDAR