{"id":552018,"date":"2018-11-20T02:39:57","date_gmt":"2018-11-20T10:39:57","guid":{"rendered":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/?post_type=msr-event&p=552018"},"modified":"2025-08-06T11:56:43","modified_gmt":"2025-08-06T18:56:43","slug":"academic-research-summit-2019-on-data-science-and-ai","status":"publish","type":"msr-event","link":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/event\/academic-research-summit-2019-on-data-science-and-ai\/","title":{"rendered":"Academic Research Summit 2019 on Data Science and AI"},"content":{"rendered":"\n\n

Venue:<\/strong> ICSR Auditorium, IIT Madras<\/p>\n

The Academic Research Summit series is a forum to foster meaningful discussion and discussion among the Indian Computer Science research community and enable collaborations to advance the state of the art and raise the bar on research efforts.<\/p>\n

The Academic Research Summit 2019 will focus on the theme of Data Science and AI<\/strong><\/p>\n

 Opens in a new tab<\/span><\/p>\n

\"\"Microsoft Research (MSR) is organising the fourth edition of the\u00a0Academic Research Summit,\u00a0in partnership with the Association for Computing Machinery (ACM) India (opens in new tab)<\/span><\/a>\u00a0and the Robert Bosch Centre for Data Science and AI at IIT Madras. (opens in new tab)<\/span><\/a> The\u00a0summit will be\u00a0held on the 24th and 25th of January 2019 at the ICSR Auditorium in IIT Madras.<\/p>\n

AI researchers\u00a0are striving to\u00a0create intelligent machines that complement human reasoning\u00a0and enrich human experiences and capabilities. At the core is the\u00a0ability to harness the explosion of digital data and computational power with advanced algorithms\u00a0that extend the ability for machines\u00a0to\u00a0learn, reason, sense, and understand\u2014enabling collaborative and natural interactions between machines and humans.\u00a0The\u00a0Academic Research Summit 2019 is focused on the theme of Data Science and AI.<\/strong><\/p>\n

The agenda\u00a0of the summit will include keynotes, talks and other tracks from distinguished researchers from India and across the world. Among the speakers expected\u00a0at the event are , Raghu Ramakrishnan (opens in new tab)<\/span><\/a>, Sunita Sarawagi (opens in new tab)<\/span><\/a>, Sriram Rajamani (opens in new tab)<\/span><\/a>,\u00a0 Amit Sharma (opens in new tab)<\/span><\/a>, Chandu Thekkath (opens in new tab)<\/span><\/a>, Varun Aggarwal (opens in new tab)<\/span><\/a> etc.<\/p>\n

More details on the list of speakers, sessions and the complete agenda will be posted here soon.<\/p>\n

Participation at the summit is via invitation and prior registration only.<\/p>\n

For any queries about the summit, please email: indiamrc@microsoft.comOpens in a new tab<\/span><\/p>\n

Broad agenda as below (subject to changes). More details will be updated soon.<\/p>\n

Thursday, 24th January 2019<\/strong><\/span><\/h2>\n

09:00 – 09:30 \u00a0 \u00a0 Registration
\n09:30 – 10:15 \u00a0 \u00a0 Welcome session<\/p>\n

10:15 – 11:15 \u00a0 \u00a0\u00a0 Plenary Talk: Raghu Ramakrishnan (opens in new tab)<\/span><\/a> , CTO for Data & Technical Fellow, Microsoft<\/p>\n

Title: Cloud + Data + ML: Opportunities and Challenges<\/strong><\/p>\n

Abstract:\u00a0<\/strong>Breakthroughs in machine learning, scale-out storage and analytics, and cost-effective cloud services are creating a fundamental shift in computing.\u00a0 Data is now a core asset.\u00a0 AI is becoming a foundational tool in realizing value from data.\u00a0 These changes offer us an opportunity to rethink business strategies and scientific enterprises in radically different, data-driven ways.<\/p>\n

\u00a0In this talk, I will discuss the current technology landscape, the opportunities to bring ML to bear on a range of data-driven tasks, and the challenges of responsible data governance when doing so on high-value and sensitive data.<\/p>\n

11:15 – 11:45 \u00a0 \u00a0\u00a0 Break<\/p>\n

11:45-\u00a0 12:45 \u00a0 \u00a0 Plenary Talk<\/strong>:\u00a0Sriram Rajamani (opens in new tab)<\/span><\/a>, Managing Director, Microsoft Research India<\/p>\n

Title:<\/b> Program Synthesis meets Machine Learning<\/strong><\/p>\n

Abstract:<\/b>\u00a0 We give a tutorial overview of program synthesis, from its first formulation by Church in 1957, through its pragmatic evolution through sketching and programing-by-examples, and \u00a0compare program synthesis with supervised machine learning. We then present our recent efforts in combining program synthesis and machine learning techniques to solve the problem of synthesizing extractors from heterogeneous data. Finally, we explore several opportunities at the intersection of program synthesis (and more broadly the PL community) and machine learning, such as pruning and ranking programs during synthesis, neural program synthesis and automatic differentiation.<\/p>\n

12:45 – 14:00\u00a0 Lunch<\/p>\n

14:00 – 15:45 \u00a0 Track: Systems Support for AI<\/b>,\u00a0 Moderated by Chandu Thekkath (opens in new tab)<\/span><\/a>, Microsoft<\/p>\n

Track Speakers: Timothy Roscoe (opens in new tab)<\/span><\/a>,\u00a0 Muthian Sivathanu (opens in new tab)<\/span><\/a> , Nishanth Chandran (opens in new tab)<\/span><\/a><\/p>\n

15:45: 16:05 \u00a0 \u00a0 Break<\/p>\n

16:05 – 16: 50\u00a0 Panel Discussion: Challenges & Opportunities in AI – some perspectives, <\/strong>Moderated by\u00a0 P J Narayanan , Director IIIT Hyderabad (opens in new tab)<\/span><\/a><\/p>\n

Panelists: Saikat Guha (opens in new tab)<\/span><\/a>, Shankar Narasimhan (opens in new tab)<\/span><\/a>,\u00a0 Ravi R (opens in new tab)<\/span><\/a>, Ganesh Sankaralingam (opens in new tab)<\/span><\/a><\/p>\n

16:50 – 18:00\u00a0 Poster Session<\/p>\n

18:00 – 19:00 Travel to Dinner <\/strong>
\n19:00 – 22:00 Dinner\u00a0<\/strong><\/p>\n

Friday, 25th January<\/strong> <\/span><\/h2>\n

09:30 – 10:30 \u00a0\u00a0 Plenary Talk:\u00a0Sunita Sarawagi (opens in new tab)<\/span><\/a>, Professor, IIT Bombay<\/strong><\/p>\n

Title: Redesigning Neural Architectures for Sequence to Sequence Learning<\/strong><\/p>\n

Abstract:<\/strong> The Encoder-Decoder model with soft-attention is now the defacto standard for sequence to sequence learning, having enjoyed early success in tasks like translation, error correction, and speech recognition.\u00a0 In this talk, I will present a critique of various aspect of this popular model, including its soft attention mechanism, local loss function, and sequential decoding.\u00a0 I will present a new Posterior Attention Network for a more transparent joint attention that provides easy gains on several translation and morphological inflection tasks.\u00a0 Next, I will expose a little known problem of mis-calibration in state of the art neural machine translation (NMT) systems. For structured outputs like in NMT, calibration is important not just for reliable confidence with predictions, but also for proper functioning of beam-search inference. I will discuss reasons for mis-calibration and some fixes.\u00a0 Finally, I will summarize recent research efforts towards parallel decoding of long sequences.<\/p>\n

10:30 – 11:20 \u00a0 \u00a0 Industry talk:<\/strong> Varun Aggarwal (opens in new tab)<\/span><\/a>, Co-Founder, Aspiring Minds<\/p>\n

Title: Using AI in the industry, sprucing up academia<\/strong><\/p>\n

Abstract<\/strong>: I will discuss how we have used AI over the last 7 years to build tools for grading skills and providing feedback. I will discuss examples of grading programs, video interviews and simulating chats. Through these, I will demonstrate the opportunities in using AI in the industry, the challenges and pitfalls. In the latter part of the talk, I will talk about the gaps in science policy in India based on my recent book. These seriously impede India’s AI efforts, which need to be immediately addressed.<\/p>\n

11:20 – 11:45 \u00a0 \u00a0\u00a0 Break<\/p>\n

11:45 – 13:45 \u00a0 \u00a0 Track: AI for Societal Impact , <\/strong>Moderated by Amit Sharma (opens in new tab)<\/span><\/a>, Microsoft Research<\/p>\n

Track Speakers: Saket Anand (opens in new tab)<\/span><\/a>, Kameswari Chebrolu (opens in new tab)<\/span><\/a>, Purvi Shah (opens in new tab)<\/span><\/a>, Ronak Sutaria (opens in new tab)<\/span><\/a><\/p>\n

13:45 – 14:30 Lunch<\/p>\n

14:30 – 15:45\u00a0 Track: Technologies for India<\/strong>, Hosted by Dr. Hemant Pande (opens in new tab)<\/span><\/a>, Executive Director, ACM India<\/p>\n

Track Speakers: Sanjay Jain (opens in new tab)<\/span><\/a> , Saurabh Panjwani (opens in new tab)<\/span><\/a> , P S Nair – TeamIndus, Siddharth Shetty – iSPIRT<\/p>\n

15:45 – 16:15\u00a0 Wrap-up & Closing<\/p>\n

 <\/p>\n

 Opens in a new tab<\/span><\/p>\n","protected":false},"excerpt":{"rendered":"

Venue: ICSR Auditorium, IIT Madras The Academic Research Summit series is a forum to foster meaningful discussion and discussion among the Indian Computer Science research community and enable collaborations to advance the state of the art and raise the bar on research efforts. The Academic Research Summit 2019 will focus on the theme of Data […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2019-01-24","msr_enddate":"2019-01-25","msr_location":"Chennai, India","msr_expirationdate":"2024-12-31","msr_event_recording_link":"","msr_event_link":"","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":false,"msr_private_event":false,"msr_hide_image_in_river":0,"footnotes":""},"research-area":[13556,13562,13563,13545],"msr-region":[197903],"msr-event-type":[197947],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-552018","msr-event","type-msr-event","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-computer-vision","msr-research-area-data-platform-analytics","msr-research-area-human-language-technologies","msr-region-asia-pacific","msr-event-type-universities","msr-locale-en_us"],"msr_about":"\n\n

Venue:<\/strong> ICSR Auditorium, IIT Madras<\/p>\n

The Academic Research Summit series is a forum to foster meaningful discussion and discussion among the Indian Computer Science research community and enable collaborations to advance the state of the art and raise the bar on research efforts.<\/p>\n

The Academic Research Summit 2019 will focus on the theme of Data Science and AI<\/strong><\/p>\n

 Opens in a new tab<\/span><\/p>\n

\"\"Microsoft Research (MSR) is organising the fourth edition of the\u00a0Academic Research Summit,\u00a0in partnership with the Association for Computing Machinery (ACM) India (opens in new tab)<\/span><\/a>\u00a0and the Robert Bosch Centre for Data Science and AI at IIT Madras. (opens in new tab)<\/span><\/a> The\u00a0summit will be\u00a0held on the 24th and 25th of January 2019 at the ICSR Auditorium in IIT Madras.<\/p>\n

AI researchers\u00a0are striving to\u00a0create intelligent machines that complement human reasoning\u00a0and enrich human experiences and capabilities. At the core is the\u00a0ability to harness the explosion of digital data and computational power with advanced algorithms\u00a0that extend the ability for machines\u00a0to\u00a0learn, reason, sense, and understand\u2014enabling collaborative and natural interactions between machines and humans.\u00a0The\u00a0Academic Research Summit 2019 is focused on the theme of Data Science and AI.<\/strong><\/p>\n

The agenda\u00a0of the summit will include keynotes, talks and other tracks from distinguished researchers from India and across the world. Among the speakers expected\u00a0at the event are , Raghu Ramakrishnan (opens in new tab)<\/span><\/a>, Sunita Sarawagi (opens in new tab)<\/span><\/a>, Sriram Rajamani (opens in new tab)<\/span><\/a>,\u00a0 Amit Sharma (opens in new tab)<\/span><\/a>, Chandu Thekkath (opens in new tab)<\/span><\/a>, Varun Aggarwal (opens in new tab)<\/span><\/a> etc.<\/p>\n

More details on the list of speakers, sessions and the complete agenda will be posted here soon.<\/p>\n

Participation at the summit is via invitation and prior registration only.<\/p>\n

For any queries about the summit, please email: indiamrc@microsoft.comOpens in a new tab<\/span><\/p>\n

Broad agenda as below (subject to changes). More details will be updated soon.<\/p>\n

Thursday, 24th January 2019<\/strong><\/span><\/h2>\n

09:00 – 09:30 \u00a0 \u00a0 Registration
\n09:30 – 10:15 \u00a0 \u00a0 Welcome session<\/p>\n

10:15 – 11:15 \u00a0 \u00a0\u00a0 Plenary Talk: Raghu Ramakrishnan (opens in new tab)<\/span><\/a> , CTO for Data & Technical Fellow, Microsoft<\/p>\n

Title: Cloud + Data + ML: Opportunities and Challenges<\/strong><\/p>\n

Abstract:\u00a0<\/strong>Breakthroughs in machine learning, scale-out storage and analytics, and cost-effective cloud services are creating a fundamental shift in computing.\u00a0 Data is now a core asset.\u00a0 AI is becoming a foundational tool in realizing value from data.\u00a0 These changes offer us an opportunity to rethink business strategies and scientific enterprises in radically different, data-driven ways.<\/p>\n

\u00a0In this talk, I will discuss the current technology landscape, the opportunities to bring ML to bear on a range of data-driven tasks, and the challenges of responsible data governance when doing so on high-value and sensitive data.<\/p>\n

11:15 – 11:45 \u00a0 \u00a0\u00a0 Break<\/p>\n

11:45-\u00a0 12:45 \u00a0 \u00a0 Plenary Talk<\/strong>:\u00a0Sriram Rajamani (opens in new tab)<\/span><\/a>, Managing Director, Microsoft Research India<\/p>\n

Title:<\/b> Program Synthesis meets Machine Learning<\/strong><\/p>\n

Abstract:<\/b>\u00a0 We give a tutorial overview of program synthesis, from its first formulation by Church in 1957, through its pragmatic evolution through sketching and programing-by-examples, and \u00a0compare program synthesis with supervised machine learning. We then present our recent efforts in combining program synthesis and machine learning techniques to solve the problem of synthesizing extractors from heterogeneous data. Finally, we explore several opportunities at the intersection of program synthesis (and more broadly the PL community) and machine learning, such as pruning and ranking programs during synthesis, neural program synthesis and automatic differentiation.<\/p>\n

12:45 – 14:00\u00a0 Lunch<\/p>\n

14:00 – 15:45 \u00a0 Track: Systems Support for AI<\/b>,\u00a0 Moderated by Chandu Thekkath (opens in new tab)<\/span><\/a>, Microsoft<\/p>\n

Track Speakers: Timothy Roscoe (opens in new tab)<\/span><\/a>,\u00a0 Muthian Sivathanu (opens in new tab)<\/span><\/a> , Nishanth Chandran (opens in new tab)<\/span><\/a><\/p>\n

15:45: 16:05 \u00a0 \u00a0 Break<\/p>\n

16:05 – 16: 50\u00a0 Panel Discussion: Challenges & Opportunities in AI – some perspectives, <\/strong>Moderated by\u00a0 P J Narayanan , Director IIIT Hyderabad (opens in new tab)<\/span><\/a><\/p>\n

Panelists: Saikat Guha (opens in new tab)<\/span><\/a>, Shankar Narasimhan (opens in new tab)<\/span><\/a>,\u00a0 Ravi R (opens in new tab)<\/span><\/a>, Ganesh Sankaralingam (opens in new tab)<\/span><\/a><\/p>\n

16:50 – 18:00\u00a0 Poster Session<\/p>\n

18:00 – 19:00 Travel to Dinner <\/strong>
\n19:00 – 22:00 Dinner\u00a0<\/strong><\/p>\n

Friday, 25th January<\/strong> <\/span><\/h2>\n

09:30 – 10:30 \u00a0\u00a0 Plenary Talk:\u00a0Sunita Sarawagi (opens in new tab)<\/span><\/a>, Professor, IIT Bombay<\/strong><\/p>\n

Title: Redesigning Neural Architectures for Sequence to Sequence Learning<\/strong><\/p>\n

Abstract:<\/strong> The Encoder-Decoder model with soft-attention is now the defacto standard for sequence to sequence learning, having enjoyed early success in tasks like translation, error correction, and speech recognition.\u00a0 In this talk, I will present a critique of various aspect of this popular model, including its soft attention mechanism, local loss function, and sequential decoding.\u00a0 I will present a new Posterior Attention Network for a more transparent joint attention that provides easy gains on several translation and morphological inflection tasks.\u00a0 Next, I will expose a little known problem of mis-calibration in state of the art neural machine translation (NMT) systems. For structured outputs like in NMT, calibration is important not just for reliable confidence with predictions, but also for proper functioning of beam-search inference. I will discuss reasons for mis-calibration and some fixes.\u00a0 Finally, I will summarize recent research efforts towards parallel decoding of long sequences.<\/p>\n

10:30 – 11:20 \u00a0 \u00a0 Industry talk:<\/strong> Varun Aggarwal (opens in new tab)<\/span><\/a>, Co-Founder, Aspiring Minds<\/p>\n

Title: Using AI in the industry, sprucing up academia<\/strong><\/p>\n

Abstract<\/strong>: I will discuss how we have used AI over the last 7 years to build tools for grading skills and providing feedback. I will discuss examples of grading programs, video interviews and simulating chats. Through these, I will demonstrate the opportunities in using AI in the industry, the challenges and pitfalls. In the latter part of the talk, I will talk about the gaps in science policy in India based on my recent book. These seriously impede India’s AI efforts, which need to be immediately addressed.<\/p>\n

11:20 – 11:45 \u00a0 \u00a0\u00a0 Break<\/p>\n

11:45 – 13:45 \u00a0 \u00a0 Track: AI for Societal Impact , <\/strong>Moderated by Amit Sharma (opens in new tab)<\/span><\/a>, Microsoft Research<\/p>\n

Track Speakers: Saket Anand (opens in new tab)<\/span><\/a>, Kameswari Chebrolu (opens in new tab)<\/span><\/a>, Purvi Shah (opens in new tab)<\/span><\/a>, Ronak Sutaria (opens in new tab)<\/span><\/a><\/p>\n

13:45 – 14:30 Lunch<\/p>\n

14:30 – 15:45\u00a0 Track: Technologies for India<\/strong>, Hosted by Dr. Hemant Pande (opens in new tab)<\/span><\/a>, Executive Director, ACM India<\/p>\n

Track Speakers: Sanjay Jain (opens in new tab)<\/span><\/a> , Saurabh Panjwani (opens in new tab)<\/span><\/a> , P S Nair – TeamIndus, Siddharth Shetty – iSPIRT<\/p>\n

15:45 – 16:15\u00a0 Wrap-up & Closing<\/p>\n

 <\/p>\n

 Opens in a new tab<\/span><\/p>\n","tab-content":[{"id":0,"name":"About","content":"\"\"Microsoft Research (MSR) is organising the fourth edition of the\u00a0Academic Research Summit,\u00a0in partnership with the Association for Computing Machinery (ACM) India<\/a>\u00a0and the Robert Bosch Centre for Data Science and AI at IIT Madras.<\/a> The\u00a0summit will be\u00a0held on the 24th and 25th of January 2019 at the ICSR Auditorium in IIT Madras.\r\n\r\nAI researchers\u00a0are striving to\u00a0create intelligent machines that complement human reasoning\u00a0and enrich human experiences and capabilities. At the core is the\u00a0ability to harness the explosion of digital data and computational power with advanced algorithms\u00a0that extend the ability for machines\u00a0to\u00a0learn, reason, sense, and understand\u2014enabling collaborative and natural interactions between machines and humans.\u00a0The\u00a0Academic Research Summit 2019 is focused on the theme of Data Science and AI.<\/strong>\r\n\r\nThe agenda\u00a0of the summit will include keynotes, talks and other tracks from distinguished researchers from India and across the world. Among the speakers expected\u00a0at the event are , Raghu Ramakrishnan<\/a>, Sunita Sarawagi<\/a>, Sriram Rajamani<\/a>,\u00a0 Amit Sharma<\/a>, Chandu Thekkath<\/a>, Varun Aggarwal<\/a> etc.\r\n\r\nMore details on the list of speakers, sessions and the complete agenda will be posted here soon.\r\n\r\nParticipation at the summit is via invitation and prior registration only.\r\n\r\nFor any queries about the summit, please email: indiamrc@microsoft.com"},{"id":1,"name":"Agenda","content":"Broad agenda as below (subject to changes). More details will be updated soon.\r\n

Thursday, 24th January 2019<\/strong><\/span><\/h2>\r\n09:00 - 09:30 \u00a0 \u00a0 Registration\r\n09:30 - 10:15 \u00a0 \u00a0 Welcome session\r\n\r\n10:15 - 11:15 \u00a0 \u00a0\u00a0 Plenary Talk: Raghu Ramakrishnan<\/a> , CTO for Data & Technical Fellow, Microsoft\r\n\r\nTitle: Cloud + Data + ML: Opportunities and Challenges<\/strong>\r\n

Abstract:\u00a0<\/strong>Breakthroughs in machine learning, scale-out storage and analytics, and cost-effective cloud services are creating a fundamental shift in computing.\u00a0 Data is now a core asset.\u00a0 AI is becoming a foundational tool in realizing value from data.\u00a0 These changes offer us an opportunity to rethink business strategies and scientific enterprises in radically different, data-driven ways.<\/p>\r\n

\u00a0In this talk, I will discuss the current technology landscape, the opportunities to bring ML to bear on a range of data-driven tasks, and the challenges of responsible data governance when doing so on high-value and sensitive data.<\/p>\r\n11:15 - 11:45 \u00a0 \u00a0\u00a0 Break\r\n\r\n11:45-\u00a0 12:45 \u00a0 \u00a0 Plenary Talk<\/strong>:\u00a0Sriram Rajamani<\/a>, Managing Director, Microsoft Research India\r\n\r\nTitle:<\/b> Program Synthesis meets Machine Learning<\/strong>\r\n

Abstract:<\/b>\u00a0 We give a tutorial overview of program synthesis, from its first formulation by Church in 1957, through its pragmatic evolution through sketching and programing-by-examples, and \u00a0compare program synthesis with supervised machine learning. We then present our recent efforts in combining program synthesis and machine learning techniques to solve the problem of synthesizing extractors from heterogeneous data. Finally, we explore several opportunities at the intersection of program synthesis (and more broadly the PL community) and machine learning, such as pruning and ranking programs during synthesis, neural program synthesis and automatic differentiation.<\/p>\r\n12:45 - 14:00\u00a0 Lunch\r\n\r\n14:00 - 15:45 \u00a0 Track: Systems Support for AI<\/b>,\u00a0 Moderated by Chandu Thekkath<\/a>, Microsoft\r\n\r\nTrack Speakers: Timothy Roscoe<\/a>,\u00a0 Muthian Sivathanu<\/a> , Nishanth Chandran<\/a>\r\n\r\n15:45: 16:05 \u00a0 \u00a0 Break\r\n\r\n16:05 - 16: 50\u00a0 Panel Discussion: Challenges & Opportunities in AI - some perspectives, <\/strong>Moderated by\u00a0 P J Narayanan , Director IIIT Hyderabad<\/a>\r\n\r\nPanelists: Saikat Guha<\/a>, Shankar Narasimhan<\/a>,\u00a0 Ravi R<\/a>, Ganesh Sankaralingam<\/a>\r\n\r\n16:50 - 18:00\u00a0 Poster Session\r\n\r\n18:00 - 19:00 Travel to Dinner <\/strong>\r\n19:00 - 22:00 Dinner\u00a0<\/strong>\r\n

Friday, 25th January<\/strong> <\/span><\/h2>\r\n09:30 - 10:30 \u00a0\u00a0 Plenary Talk:\u00a0Sunita Sarawagi<\/a>, Professor, IIT Bombay<\/strong>\r\n\r\nTitle: Redesigning Neural Architectures for Sequence to Sequence Learning<\/strong>\r\n\r\nAbstract:<\/strong> The Encoder-Decoder model with soft-attention is now the defacto standard for sequence to sequence learning, having enjoyed early success in tasks like translation, error correction, and speech recognition.\u00a0 In this talk, I will present a critique of various aspect of this popular model, including its soft attention mechanism, local loss function, and sequential decoding.\u00a0 I will present a new Posterior Attention Network for a more transparent joint attention that provides easy gains on several translation and morphological inflection tasks.\u00a0 Next, I will expose a little known problem of mis-calibration in state of the art neural machine translation (NMT) systems. For structured outputs like in NMT, calibration is important not just for reliable confidence with predictions, but also for proper functioning of beam-search inference. I will discuss reasons for mis-calibration and some fixes.\u00a0 Finally, I will summarize recent research efforts towards parallel decoding of long sequences.\r\n\r\n10:30 - 11:20 \u00a0 \u00a0 Industry talk:<\/strong> Varun Aggarwal<\/a>, Co-Founder, Aspiring Minds\r\n\r\nTitle: Using AI in the industry, sprucing up academia<\/strong>\r\n

Abstract<\/strong>: I will discuss how we have used AI over the last 7 years to build tools for grading skills and providing feedback. I will discuss examples of grading programs, video interviews and simulating chats. Through these, I will demonstrate the opportunities in using AI in the industry, the challenges and pitfalls. In the latter part of the talk, I will talk about the gaps in science policy in India based on my recent book. These seriously impede India's AI efforts, which need to be immediately addressed.<\/p>\r\n11:20 - 11:45 \u00a0 \u00a0\u00a0 Break\r\n\r\n11:45 - 13:45 \u00a0 \u00a0 Track: AI for Societal Impact , <\/strong>Moderated by Amit Sharma<\/a>, Microsoft Research\r\n\r\nTrack Speakers: Saket Anand<\/a>, Kameswari Chebrolu<\/a>, Purvi Shah<\/a>, Ronak Sutaria<\/a>\r\n\r\n13:45 - 14:30 Lunch\r\n\r\n14:30 - 15:45\u00a0 Track: Technologies for India<\/strong>, Hosted by Dr. Hemant Pande<\/a>, Executive Director, ACM India\r\n\r\nTrack Speakers: Sanjay Jain<\/a> , Saurabh Panjwani<\/a> , P S Nair - TeamIndus, Siddharth Shetty - iSPIRT\r\n\r\n15:45 - 16:15\u00a0 Wrap-up & Closing\r\n\r\n \r\n\r\n "}],"msr_startdate":"2019-01-24","msr_enddate":"2019-01-25","msr_event_time":"","msr_location":"Chennai, India","msr_event_link":"","msr_event_recording_link":"","msr_startdate_formatted":"January 24, 2019","msr_register_text":"Watch now","msr_cta_link":"","msr_cta_text":"","msr_cta_bi_name":"","featured_image_thumbnail":null,"event_excerpt":"Microsoft Research (MSR) is organising the fourth edition of the\u00a0Academic Research Summit,\u00a0in partnership with the Association for Computing Machinery (ACM) India\u00a0and the Robert Bosch Centre for Data Science and AI at IIT Madras. The\u00a0summit will be\u00a0held on the 24th and 25th of January 2019 at the ICSR Auditorium in IIT Madras. AI researchers\u00a0are striving to\u00a0create intelligent machines that complement human reasoning\u00a0and enrich human experiences and capabilities. At the core is the\u00a0ability to harness the explosion…","msr_research_lab":[199562],"related-researchers":[],"msr_impact_theme":[],"related-academic-programs":[],"related-groups":[],"related-projects":[],"related-opportunities":[],"related-publications":[],"related-videos":[572295,573228,573237,573246,573255,575787,575799,575808,575826,575835,575847,575862,575871,575880,575889,575898,575916],"related-posts":[],"_links":{"self":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/552018","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-event"}],"about":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-event"}],"version-history":[{"count":4,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/552018\/revisions"}],"predecessor-version":[{"id":1147064,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/552018\/revisions\/1147064"}],"wp:attachment":[{"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/media?parent=552018"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=552018"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=552018"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=552018"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=552018"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=552018"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=552018"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=552018"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/cm-edgetun.pages.dev\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=552018"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}