BEGIN:VCALENDAR VERSION:2.0 PRODID:IEEE vTools.Events//EN CALSCALE:GREGORIAN BEGIN:VTIMEZONE TZID:US/Eastern BEGIN:DAYLIGHT DTSTART:20210314T030000 TZOFFSETFROM:-0500 TZOFFSETTO:-0400 RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3 TZNAME:EDT END:DAYLIGHT BEGIN:STANDARD DTSTART:20211107T010000 TZOFFSETFROM:-0400 TZOFFSETTO:-0500 RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11 TZNAME:EST END:STANDARD END:VTIMEZONE BEGIN:VEVENT DTSTAMP:20220319T021901Z UID:20A1349B-88E8-49F2-A9C5-CBEEA995044D DTSTART;TZID=US/Eastern:20210412T150000 DTEND;TZID=US/Eastern:20210412T160000 DESCRIPTION:Dr. Jianfeng Gao\, Distinguished Scientist & Vice President at Microsoft\n\nTitle: Robust Conversational AI with Grounded Text Generation \n\nDate: April 12\, 2021\n\nTime: 2:00 - 3:00 pm\n\nMeeting URL: [https: //queensu.zoom.us/j/99659158948?pwd=VFVKcnIyczUyN0lxZEpxb3NSV0ZDZz09](http s://can01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fqueensu.zoom .us%2Fj%2F99659158948%3Fpwd%3DVFVKcnIyczUyN0lxZEpxb3NSV0ZDZz09&data=04%7C0 1%7Csuzan.eren%40queensu.ca%7C967bcc1442d548e0946c08d8faa55e3b%7Cd61ecb3b3 8b142d582c4efb2838b925c%7C1%7C0%7C637534935445674599%7CUnknown%7CTWFpbGZsb 3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C10 00&sdata=Gnq4OuhQKy%2FMh%2Fla2Jf8G49%2BD1HD2T03dpsXY91kWAc%3D&reserved=0)\ n\nMeeting ID: 996 5915 8948\n\nPasscode: Ingenuity\n\nAbstract:\n\nIn thi s talk\, I present a hybrid approach based on a Grounded Text Generation ( GTG) model to building robust task bots at scale. GTG is a hybrid model wh ich uses a large-scale Transform neural network as its backbone\, combined with symbol manipulation modules for knowledge base inference and prior k nowledge encoding\, to generate responses grounded in dialog belief state and real-world knowledge for task completion. GTG is pre-trained on large amounts of raw text and human conversational data\, and can be fine-tuned to complete a wide range of tasks.\n\nThe hybrid approach and its variants are being developed simultaneously by multiple research teams. The primar y results reported on task-oriented dialog benchmarks are very promising\, demonstrating big potential of this approach. I provide an overview of th is progress and discuss related methods and technologies that can be incor porated for building robust conversational AI systems.\n\nBio:\n\nJianfeng Gao is a Distinguished Scientist and Vice President of Microsoft. He is t he manager of the Deep Learning (DL) group at Microsoft Research\, leading the development of AI systems for natural language processing\, Web searc h\, vision language understanding\, dialogue\, and business applications. He is an IEEE fellow.\n\nFrom 2014 to 2017\, he was Partner Research Manag er at Deep Learning Technology Center at Microsoft Research\, Redmond\, wh ere he was leading the research on deep learning for text and image proces sing. From 2006 to 2014\, he was Principal Researcher at Natural Language Processing Group at Microsoft Research\, Redmond\, where he worked on Web search\, query understanding and reformulation\, ads prediction\, and stat istical machine translation. From 2005 to 2006\, he was a Research Lead in Natural Interactive Services Division at Microsoft\, where he worked on P roject X\, an effort of developing natural user interface for Windows. Fro m 2000 to 2005\, he was Research Lead in Natural Language Computing Group at Microsoft Research Asia\, where he and his colleagues developed the fir st Chinese speech recognition system released with Microsoft Office\, the Chinese/Japanese Input Method Editors (IME) which were the leading product s in the market\, and the natural language platform for Microsoft Windows. \n\nMore information about Dr Gao can be found here:\n\n[https://www.micro soft.com/en-us/research/people/jfgao/](https://can01.safelinks.protection. outlook.com/?url=https%3A%2F%2Fwww.microsoft.com%2Fen-us%2Fresearch%2Fpeop le%2Fjfgao%2F&data=04%7C01%7Csuzan.eren%40queensu.ca%7C967bcc1442d548e0946 c08d8faa55e3b%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C637534935445684 596%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik 1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=71s5DB55oR2uxut9xIRL6fWuL63PoV27jZZHmzG% 2Bf5w%3D&reserved=0)\n\nVirtual: https://events.vtools.ieee.org/m/269264 LOCATION:Virtual: https://events.vtools.ieee.org/m/269264 ORGANIZER:suzan.eren@queensu.ca SEQUENCE:3 SUMMARY:Robust Conversational AI with Grounded Text Generation URL;VALUE=URI:https://events.vtools.ieee.org/m/269264 X-ALT-DESC:Description: <br /><p><strong>Dr. Jianfeng Gao\, Distinguished S cientist &amp\; Vice President at Microsoft</strong></p>\n<div>\n<p><stron g>Title: Robust Conversational AI with Grounded Text Generation</strong></ p>\n<p><strong>Date:&nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\ ; &nbsp\; &nbsp\; &nbsp\; April 12\, 2021</strong></p>\n<p><strong>Time:&n bsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbsp\; &nbs p\;2:00 - 3:00 pm</strong></p>\n<p><strong>Meeting URL:&nbsp\;</strong>&nb sp\;&nbsp\;&nbsp\;&nbsp\;<a title="Original URL: https://queensu.zoom.us/j /99659158948?pwd=VFVKcnIyczUyN0lxZEpxb3NSV0ZDZz09. Click or tap if you tru st this link." href="https://can01.safelinks.protection.outlook.com/?url=h ttps%3A%2F%2Fqueensu.zoom.us%2Fj%2F99659158948%3Fpwd%3DVFVKcnIyczUyN0lxZEp xb3NSV0ZDZz09&amp\;data=04%7C01%7Csuzan.eren%40queensu.ca%7C967bcc1442d548 e0946c08d8faa55e3b%7Cd61ecb3b38b142d582c4efb2838b925c%7C1%7C0%7C6375349354 45674599%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBT iI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp\;sdata=Gnq4OuhQKy%2FMh%2Fla2Jf8G49%2B D1HD2T03dpsXY91kWAc%3D&amp\;reserved=0" target="_blank" rel="noopener nore ferrer" data-auth="Verified" data-linkindex="0">https://queensu.zoom.us/j/ 99659158948?pwd=VFVKcnIyczUyN0lxZEpxb3NSV0ZDZz09</a></p>\n<p>Meeting ID: & nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\; 996 5915 8948</p>\n<p>Passcode:& nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp\;&nbsp \; Ingenuity</p>\n</div>\n<p><strong><span lang="en-US">Abstract:</span></ strong></p>\n<p><span lang="en-US">In this talk\, I present a hybrid appro ach based on a Grounded Text Generation (GTG) model to building robust tas k bots at scale.&nbsp\; GTG is a hybrid model which uses a large-scale Tra nsform neural network as its backbone\, combined with symbol manipulation modules for knowledge base inference and prior knowledge encoding\, to gen erate responses grounded in dialog belief state and real-world knowledge f or task completion.&nbsp\; GTG is pre-trained on large amounts of raw text and human conversational data\, and can be fine-tuned to complete a wide range of tasks.&nbsp\;</span></p>\n<p><span lang="en-US">The hybrid approa ch and its variants are being developed simultaneously by multiple researc h teams. The primary results reported on task-oriented dialog benchmarks a re very promising\, demonstrating big potential of this approach. I provid e an overview of this progress and discuss related methods and technologie s that can be incorporated for building robust conversational AI systems.< /span></p>\n<p><span lang="en-US">&nbsp\;</span></p>\n<p><strong><span lan g="en-US">Bio: &nbsp\;</span></strong></p>\n<p><strong><span lang="en-US"> Jianfeng Gao</span></strong><span lang="en-US">&nbsp\;is a Distinguished S cientist and Vice President of Microsoft. He is the manager of the Deep Le arning (DL) group at Microsoft Research\, leading the development of AI sy stems for natural language processing\, Web search\, vision language under standing\, dialogue\, and business applications. He</span><span lang="en-U S">&nbsp\;is an IEEE fellow.</span></p>\n<p><span lang="en-US">&nbsp\;</sp an></p>\n<p><span lang="en-US">From 2014 to 2017\, he was Partner Research Manager at Deep Learning Technology Center at Microsoft Research\, Redmon d\, where he was leading the research on deep learning for text and image processing. From 2006 to 2014\, he was Principal Researcher at Natural Lan guage Processing Group at Microsoft Research\, Redmond\, where he worked o n Web search\, query understanding and reformulation\, ads prediction\, an d statistical machine translation. From 2005 to 2006\, he was a Research L ead in Natural Interactive Services Division at Microsoft\, where he worke d on Project X\, an effort of developing natural user interface for Window s. From 2000 to 2005\, he was Research Lead in Natural Language Computing Group at Microsoft Research Asia\, where he and his colleagues developed t he first Chinese speech recognition system released with Microsoft Office\ , the Chinese/Japanese Input Method Editors (IME) which were the leading p roducts in the market\, and the natural language platform for Microsoft Wi ndows.</span></p>\n<p>More information about Dr Gao can be found here:</p> \n<p><a title="Original URL: https://www.microsoft.com/en-us/research/peop le/jfgao/. Click or tap if you trust this link." href="https://can01.safel inks.protection.outlook.com/?url=https%3A%2F%2Fwww.microsoft.com%2Fen-us%2 Fresearch%2Fpeople%2Fjfgao%2F&amp\;data=04%7C01%7Csuzan.eren%40queensu.ca% 7C967bcc1442d548e0946c08d8faa55e3b%7Cd61ecb3b38b142d582c4efb2838b925c%7C1% 7C0%7C637534935445684596%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQI joiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&amp\;sdata=71s5DB55oR2uxu t9xIRL6fWuL63PoV27jZZHmzG%2Bf5w%3D&amp\;reserved=0" target="_blank" rel="n oopener noreferrer" data-auth="Verified" data-linkindex="1">https://www.mi crosoft.com/en-us/research/people/jfgao/</a></p> END:VEVENT END:VCALENDAR