From yano at cs.cmu.edu Wed Feb 4 13:12:24 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Wed, 4 Feb 2015 13:12:24 -0500 Subject: [Intelligence Seminar] AI Seminar: Francesca Rossi Message-ID: Please join us for our first AI Seminar of 2015! Tuesday, February 24, 2015, at 3:30pm in GHC 6115. Francesca Rossi University of Padova, Italy What AI can do for multi-item sentiment analysis Abstract: Sentiment analysis assigns a positive, negative or neutral polarity to an item or entity, extracting and aggregating individual opinions from their textual expressions by means of natural language processing tools. It then aggregates the individuals' opinions into a collective sentiment about the item under consideration. Current sentiment analysis techniques are satisfactory in case there is a single entity under consideration, but can lead to inaccurate or wrong results when dealing with a set of possibly correlated items. This can be useful, for example, when a company wants to know the collective preference order over a set of products, or when we want to predict the outcome of an election over a collection of candidates. We argue that, in order to deal with this more general setting, we should exploit AI techniques such as those used in preference reasoning, multi-agent systems, and computational social choice. Preference modelling and reasoning tools provide the useful ingredients to model individual's preferences in the most faithful way, while computational social choice techniques give methods to aggregate such preferences which satisfy certain desired properties. Other AI techniques can be very useful as well, such as machine learning or recommender system tools to cope with incompleteness in the information provided by each individual. We describe a social choice aggregation rule which combines individuals' sentiment and preference information. We show that this rule satisfies a number of properties which have a natural interpretation in the sentiment analysis domain, and we evaluate its behavior when faced with highly incomplete domains. Research done in collaboration with U. Grandi, A. Loreggia, and V. Saraswat Bio: Francesca Rossi is a professor of computer science at the University of Padova, Italy. Currently she is on sabbatical at Harvard with a fellowship of the Radcliffe Institute for Advanced Studiy. Her research interests include: constraint reasoning, preferences, multi-agent systems, computational social choice and artificial intelligence. She has been president of the international association for constraint programming (ACP) and is now the president of IJCAI. She has been program chair of CP 2003 and of IJCAI 2013. She is on the editorial board of Constraints, Artificial Intelligence, AMAI and KAIS, and is Associate Editor in Chief of JAIR. She has published over 160 articles in international journals, proceedings of international conferences or workshops, and as book chapters. She has co-authored a book, edited 16 volumes of conference proceedings, collections of contributions, and special issue of international journals, and has co-edited the Handbook of Constraint Programming. Faculty Host: Tuomas Sandholm For appointments, please contact Charlotte Yano (yano at cs.cmu.edu) ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Francesca Rossi 2015.pdf Type: application/pdf Size: 491569 bytes Desc: AI Seminar Francesca Rossi 2015.pdf URL: From yano at cs.cmu.edu Mon Feb 23 15:55:05 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Mon, 23 Feb 2015 15:55:05 -0500 Subject: [AI Seminar] REMINDER: AI Seminar: Francesca Rossi Message-ID: Don't forget to join us for our first AI Seminar of 2015! Tuesday, February 24, 2015, at 3:30pm in GHC 6115. Francesca Rossi University of Padova, Italy What AI can do for multi-item sentiment analysis Abstract: Sentiment analysis assigns a positive, negative or neutral polarity to an item or entity, extracting and aggregating individual opinions from their textual expressions by means of natural language processing tools. It then aggregates the individuals' opinions into a collective sentiment about the item under consideration. Current sentiment analysis techniques are satisfactory in case there is a single entity under consideration, but can lead to inaccurate or wrong results when dealing with a set of possibly correlated items. This can be useful, for example, when a company wants to know the collective preference order over a set of products, or when we want to predict the outcome of an election over a collection of candidates. We argue that, in order to deal with this more general setting, we should exploit AI techniques such as those used in preference reasoning, multi-agent systems, and computational social choice. Preference modelling and reasoning tools provide the useful ingredients to model individual's preferences in the most faithful way, while computational social choice techniques give methods to aggregate such preferences which satisfy certain desired properties. Other AI techniques can be very useful as well, such as machine learning or recommender system tools to cope with incompleteness in the information provided by each individual. We describe a social choice aggregation rule which combines individuals' sentiment and preference information. We show that this rule satisfies a number of properties which have a natural interpretation in the sentiment analysis domain, and we evaluate its behavior when faced with highly incomplete domains. Research done in collaboration with U. Grandi, A. Loreggia, and V. Saraswat Bio: Francesca Rossi is a professor of computer science at the University of Padova, Italy. Currently she is on sabbatical at Harvard with a fellowship of the Radcliffe Institute for Advanced Studiy. Her research interests include: constraint reasoning, preferences, multi-agent systems, computational social choice and artificial intelligence. She has been president of the international association for constraint programming (ACP) and is now the president of IJCAI. She has been program chair of CP 2003 and of IJCAI 2013. She is on the editorial board of Constraints, Artificial Intelligence, AMAI and KAIS, and is Associate Editor in Chief of JAIR. She has published over 160 articles in international journals, proceedings of international conferences or workshops, and as book chapters. She has co-authored a book, edited 16 volumes of conference proceedings, collections of contributions, and special issue of international journals, and has co-edited the Handbook of Constraint Programming. Faculty Host: Tuomas Sandholm For appointments, please contact Charlotte Yano (yano at cs.cmu.edu) ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Francesca Rossi 2015.pdf Type: application/pdf Size: 491569 bytes Desc: AI Seminar Francesca Rossi 2015.pdf URL: From peterb at pitt.edu Mon Feb 23 16:34:36 2015 From: peterb at pitt.edu (Peter Brusilovsky) Date: Mon, 23 Feb 2015 16:34:36 -0500 Subject: [AI Seminar] REMINDER: AI Seminar: Francesca Rossi In-Reply-To: References: Message-ID: So far it is most popular talk on Tuesday? http://halley.exp.sis.pitt.edu/comet/calendar.do Peter Brusilovsky, Professor School of Information Sciences and Intelligent Systems Program University of Pittsburgh http://www.sis.pitt.edu/~peterb Editor-in-Chief IEEE Transactions on Learning Technologies http://www.computer.org/tlt On Feb 23, 2015, at 3:55 PM, Charlotte Yano wrote: > Don?t forget to join us for our first AI Seminar of 2015! > Tuesday, February 24, 2015, at 3:30pm in GHC 6115. > > Francesca Rossi > University of Padova, Italy > > What AI can do for multi-item sentiment analysis > > Abstract: > Sentiment analysis assigns a positive, negative or neutral polarity to an item or entity, extracting and aggregating individual opinions from their textual expressions by means of natural language processing tools. It then aggregates the individuals' opinions into a collective sentiment about the item under consideration. > > Current sentiment analysis techniques are satisfactory in case there is a single entity under consideration, but can lead to inaccurate or wrong results when dealing with a set of possibly correlated items. This can be useful, for example, when a company wants to know the collective preference order over a set of products, or when we want to predict the outcome of an election over a collection of candidates. > > We argue that, in order to deal with this more general setting, we should exploit AI techniques such as those used in preference reasoning, multi-agent systems, and computational social choice. > Preference modelling and reasoning tools provide the useful ingredients to model individual's preferences in the most faithful way, while computational social choice techniques give methods to aggregate such preferences which satisfy certain desired properties. Other AI techniques can be very useful as well, such as machine learning or recommender system tools to cope with incompleteness in the information provided by each individual. > > We describe a social choice aggregation rule which combines individuals' sentiment and preference information. We show that this rule satisfies a number of properties which have a natural interpretation in the sentiment analysis domain, and we evaluate its behavior when faced with highly incomplete domains. Research done in collaboration with U. Grandi, A. Loreggia, and V. Saraswat > > Bio: > Francesca Rossi is a professor of computer science at the University of Padova, Italy. Currently she is on sabbatical at Harvard with a fellowship of the Radcliffe Institute for Advanced Studiy. Her research interests include: constraint reasoning, preferences, multi-agent systems, computational social choice and artificial intelligence. She has been president of the international association for constraint programming (ACP) and is now the president of IJCAI. She has been program chair of CP 2003 and of IJCAI 2013. She is on the editorial board of Constraints, Artificial Intelligence, AMAI and KAIS, and is Associate Editor in Chief of JAIR. She has published over 160 articles in international journals, proceedings of international conferences or workshops, and as book chapters. She has co-authored a book, edited 16 volumes of conference proceedings, collections of contributions, and special issue of international journals, and has co-edited the Handbook of Constraint Programming. > > Faculty Host: Tuomas Sandholm > For appointments, please contact Charlotte Yano (yano at cs.cmu.edu) > > > > ___________________________________________ > > Charlotte Yano > Building Curator > Administrative Associate II > Carnegie Mellon University, Computer Science Dept. > 5000 Forbes Ave. > Pittsburgh, PA 15213 > 412-268-7656 > ___________________________________________ > cogs are for gears, not people. > ___________________________________________ > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sandholm at cs.cmu.edu Tue Feb 24 15:17:10 2015 From: sandholm at cs.cmu.edu (Tuomas Sandholm) Date: Tue, 24 Feb 2015 15:17:10 -0500 Subject: [AI Seminar] REMINDER: AI Seminar: Francesca Rossi In-Reply-To: References: Message-ID: Reminder: Francesca Rossi talk in 15 minutes. There will be coffee and cookies! Best, Tuomas ********************************************************* Tuomas Sandholm, Professor Computer Science Department (Gates Hillman Complex room 9205) Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 Executive Assistant: sandholm at cs.cmu.edu Charlotte Yano www.cs.cmu.edu/~sandholm yano at cs.cmu.edu, (412) 268-7656 Phone (412) 268-8216 ********************************************************* From: Charlotte Yano Sent: Monday, February 23, 2015 3:55 PM To: ai-seminar-announce at cs.cmu.edu Cc: Tuomas Sandholm Subject: REMINDER: AI Seminar: Francesca Rossi Don't forget to join us for our first AI Seminar of 2015! Tuesday, February 24, 2015, at 3:30pm in GHC 6115. Francesca Rossi University of Padova, Italy What AI can do for multi-item sentiment analysis Abstract: Sentiment analysis assigns a positive, negative or neutral polarity to an item or entity, extracting and aggregating individual opinions from their textual expressions by means of natural language processing tools. It then aggregates the individuals' opinions into a collective sentiment about the item under consideration. Current sentiment analysis techniques are satisfactory in case there is a single entity under consideration, but can lead to inaccurate or wrong results when dealing with a set of possibly correlated items. This can be useful, for example, when a company wants to know the collective preference order over a set of products, or when we want to predict the outcome of an election over a collection of candidates. We argue that, in order to deal with this more general setting, we should exploit AI techniques such as those used in preference reasoning, multi-agent systems, and computational social choice. Preference modelling and reasoning tools provide the useful ingredients to model individual's preferences in the most faithful way, while computational social choice techniques give methods to aggregate such preferences which satisfy certain desired properties. Other AI techniques can be very useful as well, such as machine learning or recommender system tools to cope with incompleteness in the information provided by each individual. We describe a social choice aggregation rule which combines individuals' sentiment and preference information. We show that this rule satisfies a number of properties which have a natural interpretation in the sentiment analysis domain, and we evaluate its behavior when faced with highly incomplete domains. Research done in collaboration with U. Grandi, A. Loreggia, and V. Saraswat Bio: Francesca Rossi is a professor of computer science at the University of Padova, Italy. Currently she is on sabbatical at Harvard with a fellowship of the Radcliffe Institute for Advanced Studiy. Her research interests include: constraint reasoning, preferences, multi-agent systems, computational social choice and artificial intelligence. She has been president of the international association for constraint programming (ACP) and is now the president of IJCAI. She has been program chair of CP 2003 and of IJCAI 2013. She is on the editorial board of Constraints, Artificial Intelligence, AMAI and KAIS, and is Associate Editor in Chief of JAIR. She has published over 160 articles in international journals, proceedings of international conferences or workshops, and as book chapters. She has co-authored a book, edited 16 volumes of conference proceedings, collections of contributions, and special issue of international journals, and has co-edited the Handbook of Constraint Programming. Faculty Host: Tuomas Sandholm For appointments, please contact Charlotte Yano (yano at cs.cmu.edu) ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From yano at cs.cmu.edu Tue Mar 31 11:38:47 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Tue, 31 Mar 2015 11:38:47 -0400 Subject: [AI Seminar] AI Seminar: Vince Conitzer, Apr. 14 Message-ID: Please join us for the next AI Seminar! April 14th, at 10:30am in GHC 8102 **Please note the special time and location** Vincent Conitzer Duke University Computational Social Choice: A Journey from Basic Complexity Results to a Brave New World for Social Choice abstract: Social choice concerns making collective decisions based on the preferences of multiple agents, for example by voting over the alternatives. In this talk, I will cover several topics in *computational* social choice. These topics range from problems with which the field started, such as the complexity of executing and of manipulating certain voting rules, to problems arising from new applications, such as voting in combinatorial domains and in highly anonymous environments. No previous background in social choice / voting will be assumed. bio: Vincent Conitzer is the Sally Dalton Robinson Professor of Computer Science and Professor of Economics at Duke University. He received Ph.D. (2006) and M.S. (2003) degrees in Computer Science from Carnegie Mellon University, and an A.B. (2001) degree in Applied Mathematics from Harvard University. His research focuses on computational aspects of microeconomics, in particular game theory, mechanism design, voting/social choice, and auctions. This work uses techniques from, and includes applications to, artificial intelligence and multiagent systems. Conitzer has received the Social Choice and Welfare Prize (2014), a Presidential Early Career Award for Scientists and Engineers (PECASE), the IJCAI Computers and Thought Award, an NSF CAREER award, the inaugural Victor Lesser dissertation award, an honorable mention for the ACM dissertation award, and several awards for papers and service at the AAAI and AAMAS conferences. He has also been named a Kavli Fellow, a Bass Fellow, a Sloan Fellow, and one of AI's Ten to Watch. Conitzer and Preston McAfee are the founding Editors-in-Chief of the ACM Transactions on Economics and Computation (TEAC). www.cs.duke.edu/~conitzer/ Due to travel restrictions, meetings will be *very* limited. Please contact Charlotte Yano to schedule. ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Vincent Conitzer 2015docx.pdf Type: application/pdf Size: 445794 bytes Desc: AI Seminar Vincent Conitzer 2015docx.pdf URL: From yano at cs.cmu.edu Tue Apr 14 08:47:41 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Tue, 14 Apr 2015 08:47:41 -0400 Subject: [AI Seminar] REMINDER: AI Seminar: Vince Conitzer, Apr. 14 TODAY Message-ID: Don't forget to come to the AI Seminar THIS MORNING :) Talk will begin at 10:30 am, coffee and chatting starting at 10 am Thanks, -Charlotte On 03/31/2015 11:38 AM, Charlotte Yano wrote: Please join us for the next AI Seminar! April 14th, at 10:30am in GHC 8102 **Please note the special time and location** Vincent Conitzer Duke University Computational Social Choice: A Journey from Basic Complexity Results to a Brave New World for Social Choice abstract: Social choice concerns making collective decisions based on the preferences of multiple agents, for example by voting over the alternatives. In this talk, I will cover several topics in *computational* social choice. These topics range from problems with which the field started, such as the complexity of executing and of manipulating certain voting rules, to problems arising from new applications, such as voting in combinatorial domains and in highly anonymous environments. No previous background in social choice / voting will be assumed. bio: Vincent Conitzer is the Sally Dalton Robinson Professor of Computer Science and Professor of Economics at Duke University. He received Ph.D. (2006) and M.S. (2003) degrees in Computer Science from Carnegie Mellon University, and an A.B. (2001) degree in Applied Mathematics from Harvard University. His research focuses on computational aspects of microeconomics, in particular game theory, mechanism design, voting/social choice, and auctions. This work uses techniques from, and includes applications to, artificial intelligence and multiagent systems. Conitzer has received the Social Choice and Welfare Prize (2014), a Presidential Early Career Award for Scientists and Engineers (PECASE), the IJCAI Computers and Thought Award, an NSF CAREER award, the inaugural Victor Lesser dissertation award, an honorable mention for the ACM dissertation award, and several awards for papers and service at the AAAI and AAMAS conferences. He has also been named a Kavli Fellow, a Bass Fellow, a Sloan Fellow, and one of AI's Ten to Watch. Conitzer and Preston McAfee are the founding Editors-in-Chief of the ACM Transactions on Economics and Computation (TEAC). www.cs.duke.edu/~conitzer/ Due to travel restrictions, meetings will be *very* limited. Please contact Charlotte Yano to schedule. ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From yano at cs.cmu.edu Tue Apr 21 11:13:12 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Tue, 21 Apr 2015 11:13:12 -0400 Subject: [AI Seminar] AI Seminar: Kevin Leyton-Brown, Apr. 28 Message-ID: Our next AI Seminar will be Kevin Leyton-Brown, from the University of British Columbia. The talk will be on April 28th, at 3:30 in 6115. I will send around another announcement when I have the title and abstract. He will be available for meetings on the 28th and 29th, please contact Charlotte Yano for availability. Bio: Kevin Leyton-Brown is a professor of computer science at the University of British Columbia. He holds a PhD and M.Sc. from Stanford University (2003; 2001) and a B.Sc. from McMaster University (1998). He studies the intersection of computer science and microeconomics, addressing computational problems in economic contexts and incentive issues in multiagent systems. He also applies machine learning to the automated design and analysis of algorithms for solving hard computational problems. He has co-written two books, "Multiagent Systems" and "Essentials of Game Theory," and over 100 peer-refereed technical articles; his work has received over 6,000 citations and an h-index of 33. He is the recipient of a 2014 NSERC E.W.R. Steacie Memorial Fellowship-previously given to a computer scientist only 10 times since its establishment in 1965-and a 2013 Outstanding Young Computer Science Researcher Prize from the Canadian Association of Computer Science. He and his coauthors have received paper awards from JAIR, ACM-EC, AAMAS and LION, and numerous medals for the portfolio-based SAT solver SATzilla at international SAT competitions (2003-12). He serves as an associate editor for the Artificial Intelligence Journal (AIJ), ACM Transactions on Economics and Computation (ACM-TEAC), and AI Access; serves as an advisory board member for the Journal of Artificial Intelligence Research (JAIR, after serving as associate editor for two 4-year terms), and was program chair for the ACM Conference on Electronic Commerce (ACM-EC) in 2012. He has co-taught two Coursera courses on "Game Theory" to over half a million students, and has received awards for his teaching at UBC-notably, a 2013/14 Killam Teaching Prize. He split his 2010-11 sabbatical between Makerere University in Kampala, Uganda, and the Institute for Advanced Studies at Hebrew University of Jerusalem, Israel. He currently advises Auctionomics, Inc. (and through them, the Federal Communications Commission), Zynga, Inc., and Qudos, Inc. He is a co-founder of Kudu.ug and a new UBC spinoff, Meta-Algorithmic Technologies. In the past, he served as a consultant for Trading Dynamics Inc., Ariba Inc., Cariocas Inc., and was scientific advisor to UBC spinoff Zite Inc. until it was acquired by CNN in 2011. http://www.cs.ubc.ca/~kevinlb/index.html ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: From yano at cs.cmu.edu Wed Apr 22 14:00:37 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Wed, 22 Apr 2015 14:00:37 -0400 Subject: [AI Seminar] UPDATE: AI Seminar: Kevin Leyton-Brown, Apr. 28 Message-ID: I now have the talk title and abstract. Meetings are available on the 28th and 29th. Kevin Leyton-Brown University of British Columbia. 3:30pm, Apr. 28, 2015 GHC 6115 Title: Incentive Auctions and Spectrum Repacking Abstract: This talk will discuss the FCC's upcoming "incentive auction", in which television broadcasters will be given the opportunity to sell their broadcast rights, remaining broadcasters will be repacked into a smaller block of spectrum, and the freed airwaves will be resold to telecom companies. The stakes for this auction are huge-projected tens of billions of dollars in revenue for the government-justifying the design of a special-purpose descending-price auction mechanism, which I'll discuss. An inner-loop problem in this mechanism is determining whether a given set of broadcasters can be repacked into a smaller block of spectrum while respecting radio interference constraints. This is an instance of a (worst-case intractable) graph coloring problem; however, stations' broadcast locations and interference constraints are all known in advance. Early efforts to solve this problem considered mixed-integer programming formulations, but were unable to reliably solve realistic, national-scale problem instances. We advocate instead for the use of a SAT encoding, paired with a wide range of techniques: constraint graph decomposition; novel caching mechanisms that allow reuse of partial solutions from related, solved problems; algorithm configuration; algorithm portfolios; and the marriage of local-search and complete solver strategies. Considering a set of realistic problems derived from auction simulations, we show that our approach solves virtually all within the short time budget required in practice. Bio: Kevin Leyton-Brown is a professor of computer science at the University of British Columbia. He holds a PhD and M.Sc. from Stanford University (2003; 2001) and a B.Sc. from McMaster University (1998). He studies the intersection of computer science and microeconomics, addressing computational problems in economic contexts and incentive issues in multiagent systems. He also applies machine learning to the automated design and analysis of algorithms for solving hard computational problems. He has co-written two books, "Multiagent Systems" and "Essentials of Game Theory," and over 100 peer-refereed technical articles; his work has received over 6,000 citations and an h-index of 33. He is the recipient of a 2014 NSERC E.W.R. Steacie Memorial Fellowship-previously given to a computer scientist only 10 times since its establishment in 1965-and a 2013 Outstanding Young Computer Science Researcher Prize from the Canadian Association of Computer Science. He and his coauthors have received paper awards from JAIR, ACM-EC, AAMAS and LION, and numerous medals for the portfolio-based SAT solver SATzilla at international SAT competitions (2003-12). He serves as an associate editor for the Artificial Intelligence Journal (AIJ), ACM Transactions on Economics and Computation (ACM-TEAC), and AI Access; serves as an advisory board member for the Journal of Artificial Intelligence Research (JAIR, after serving as associate editor for two 4-year terms), and was program chair for the ACM Conference on Electronic Commerce (ACM-EC) in 2012. He has co-taught two Coursera courses on "Game Theory" to over half a million students, and has received awards for his teaching at UBC-notably, a 2013/14 Killam Teaching Prize. He split his 2010-11 sabbatical between Makerere University in Kampala, Uganda, and the Institute for Advanced Studies at Hebrew University of Jerusalem, Israel. He currently advises Auctionomics, Inc. (and through them, the Federal Communications Commission), Zynga, Inc., and Qudos, Inc. He is a co-founder of Kudu.ug and a new UBC spinoff, Meta-Algorithmic Technologies. In the past, he served as a consultant for Trading Dynamics Inc., Ariba Inc., Cariocas Inc., and was scientific advisor to UBC spinoff Zite Inc. until it was acquired by CNN in 2011. http://www.cs.ubc.ca/~kevinlb/index.html ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Kevin Leyton-Brown 2015.pdf Type: application/pdf Size: 412185 bytes Desc: AI Seminar Kevin Leyton-Brown 2015.pdf URL: From yano at cs.cmu.edu Mon Apr 27 09:22:44 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Mon, 27 Apr 2015 09:22:44 -0400 Subject: [AI Seminar] REMINDER: AI Seminar: Kevin Leyton-Brown, Apr. 28 Message-ID: Don't forget, AI Seminar is tomorrow at 3:30. Meeting times are still available, please contact me if you are interested. Kevin Leyton-Brown University of British Columbia. 3:30pm, Apr. 28, 2015 GHC 6115 Title: Incentive Auctions and Spectrum Repacking Abstract: This talk will discuss the FCC's upcoming "incentive auction", in which television broadcasters will be given the opportunity to sell their broadcast rights, remaining broadcasters will be repacked into a smaller block of spectrum, and the freed airwaves will be resold to telecom companies. The stakes for this auction are huge-projected tens of billions of dollars in revenue for the government-justifying the design of a special-purpose descending-price auction mechanism, which I'll discuss. An inner-loop problem in this mechanism is determining whether a given set of broadcasters can be repacked into a smaller block of spectrum while respecting radio interference constraints. This is an instance of a (worst-case intractable) graph coloring problem; however, stations' broadcast locations and interference constraints are all known in advance. Early efforts to solve this problem considered mixed-integer programming formulations, but were unable to reliably solve realistic, national-scale problem instances. We advocate instead for the use of a SAT encoding, paired with a wide range of techniques: constraint graph decomposition; novel caching mechanisms that allow reuse of partial solutions from related, solved problems; algorithm configuration; algorithm portfolios; and the marriage of local-search and complete solver strategies. Considering a set of realistic problems derived from auction simulations, we show that our approach solves virtually all within the short time budget required in practice. Bio: Kevin Leyton-Brown is a professor of computer science at the University of British Columbia. He holds a PhD and M.Sc. from Stanford University (2003; 2001) and a B.Sc. from McMaster University (1998). He studies the intersection of computer science and microeconomics, addressing computational problems in economic contexts and incentive issues in multiagent systems. He also applies machine learning to the automated design and analysis of algorithms for solving hard computational problems. He has co-written two books, "Multiagent Systems" and "Essentials of Game Theory," and over 100 peer-refereed technical articles; his work has received over 6,000 citations and an h-index of 33. He is the recipient of a 2014 NSERC E.W.R. Steacie Memorial Fellowship-previously given to a computer scientist only 10 times since its establishment in 1965-and a 2013 Outstanding Young Computer Science Researcher Prize from the Canadian Association of Computer Science. He and his coauthors have received paper awards from JAIR, ACM-EC, AAMAS and LION, and numerous medals for the portfolio-based SAT solver SATzilla at international SAT competitions (2003-12). He serves as an associate editor for the Artificial Intelligence Journal (AIJ), ACM Transactions on Economics and Computation (ACM-TEAC), and AI Access; serves as an advisory board member for the Journal of Artificial Intelligence Research (JAIR, after serving as associate editor for two 4-year terms), and was program chair for the ACM Conference on Electronic Commerce (ACM-EC) in 2012. He has co-taught two Coursera courses on "Game Theory" to over half a million students, and has received awards for his teaching at UBC-notably, a 2013/14 Killam Teaching Prize. He split his 2010-11 sabbatical between Makerere University in Kampala, Uganda, and the Institute for Advanced Studies at Hebrew University of Jerusalem, Israel. He currently advises Auctionomics, Inc. (and through them, the Federal Communications Commission), Zynga, Inc., and Qudos, Inc. He is a co-founder of Kudu.ug and a new UBC spinoff, Meta-Algorithmic Technologies. In the past, he served as a consultant for Trading Dynamics Inc., Ariba Inc., Cariocas Inc., and was scientific advisor to UBC spinoff Zite Inc. until it was acquired by CNN in 2011. http://www.cs.ubc.ca/~kevinlb/index.html ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Kevin Leyton-Brown 2015.pdf Type: application/pdf Size: 412185 bytes Desc: AI Seminar Kevin Leyton-Brown 2015.pdf URL: From yano at cs.cmu.edu Tue Apr 28 09:42:53 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Tue, 28 Apr 2015 09:42:53 -0400 Subject: [AI Seminar] FW: REMINDER: AI Seminar: Kevin Leyton-Brown, Apr. 28 Message-ID: Please join us for the AI Seminar TODAY at 3:30. Meeting times are still available, please contact me if you are interested. Kevin Leyton-Brown University of British Columbia. 3:30pm, Apr. 28, 2015 GHC 6115 Title: Incentive Auctions and Spectrum Repacking Abstract: This talk will discuss the FCC's upcoming "incentive auction", in which television broadcasters will be given the opportunity to sell their broadcast rights, remaining broadcasters will be repacked into a smaller block of spectrum, and the freed airwaves will be resold to telecom companies. The stakes for this auction are huge-projected tens of billions of dollars in revenue for the government-justifying the design of a special-purpose descending-price auction mechanism, which I'll discuss. An inner-loop problem in this mechanism is determining whether a given set of broadcasters can be repacked into a smaller block of spectrum while respecting radio interference constraints. This is an instance of a (worst-case intractable) graph coloring problem; however, stations' broadcast locations and interference constraints are all known in advance. Early efforts to solve this problem considered mixed-integer programming formulations, but were unable to reliably solve realistic, national-scale problem instances. We advocate instead for the use of a SAT encoding, paired with a wide range of techniques: constraint graph decomposition; novel caching mechanisms that allow reuse of partial solutions from related, solved problems; algorithm configuration; algorithm portfolios; and the marriage of local-search and complete solver strategies. Considering a set of realistic problems derived from auction simulations, we show that our approach solves virtually all within the short time budget required in practice. Bio: Kevin Leyton-Brown is a professor of computer science at the University of British Columbia. He holds a PhD and M.Sc. from Stanford University (2003; 2001) and a B.Sc. from McMaster University (1998). He studies the intersection of computer science and microeconomics, addressing computational problems in economic contexts and incentive issues in multiagent systems. He also applies machine learning to the automated design and analysis of algorithms for solving hard computational problems. He has co-written two books, "Multiagent Systems" and "Essentials of Game Theory," and over 100 peer-refereed technical articles; his work has received over 6,000 citations and an h-index of 33. He is the recipient of a 2014 NSERC E.W.R. Steacie Memorial Fellowship-previously given to a computer scientist only 10 times since its establishment in 1965-and a 2013 Outstanding Young Computer Science Researcher Prize from the Canadian Association of Computer Science. He and his coauthors have received paper awards from JAIR, ACM-EC, AAMAS and LION, and numerous medals for the portfolio-based SAT solver SATzilla at international SAT competitions (2003-12). He serves as an associate editor for the Artificial Intelligence Journal (AIJ), ACM Transactions on Economics and Computation (ACM-TEAC), and AI Access; serves as an advisory board member for the Journal of Artificial Intelligence Research (JAIR, after serving as associate editor for two 4-year terms), and was program chair for the ACM Conference on Electronic Commerce (ACM-EC) in 2012. He has co-taught two Coursera courses on "Game Theory" to over half a million students, and has received awards for his teaching at UBC-notably, a 2013/14 Killam Teaching Prize. He split his 2010-11 sabbatical between Makerere University in Kampala, Uganda, and the Institute for Advanced Studies at Hebrew University of Jerusalem, Israel. He currently advises Auctionomics, Inc. (and through them, the Federal Communications Commission), Zynga, Inc., and Qudos, Inc. He is a co-founder of Kudu.ug and a new UBC spinoff, Meta-Algorithmic Technologies. In the past, he served as a consultant for Trading Dynamics Inc., Ariba Inc., Cariocas Inc., and was scientific advisor to UBC spinoff Zite Inc. until it was acquired by CNN in 2011. http://www.cs.ubc.ca/~kevinlb/index.html ___________________________________________ Charlotte Yano Building Curator Administrative Associate II Carnegie Mellon University, Computer Science Dept. 5000 Forbes Ave. Pittsburgh, PA 15213 412-268-7656 ___________________________________________ cogs are for gears, not people. ___________________________________________ -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: AI Seminar Kevin Leyton-Brown 2015.pdf Type: application/pdf Size: 412185 bytes Desc: AI Seminar Kevin Leyton-Brown 2015.pdf URL: From ebrun at cs.cmu.edu Wed Apr 29 11:53:10 2015 From: ebrun at cs.cmu.edu (Emma Brunskill) Date: Wed, 29 Apr 2015 11:53:10 -0400 Subject: [AI Seminar] grad AI final project presentations Message-ID: Hi everyone, Graduate AI is having their final project presentations in the 7th floor atrium (over by the windows, near the open wood staircase) of Gates Hall from noon-1:30pm today. Please swing by if interested! Emma P.S. Apologies if this is a misuse of the list. Emma Brunskill Assistant Professor Computer Science Dept CMU http://cs.cmu.edu/~ebrun From sdinardo at cs.cmu.edu Tue Jul 14 11:38:15 2015 From: sdinardo at cs.cmu.edu (Samantha J Dinardo) Date: Tue, 14 Jul 2015 11:38:15 -0400 Subject: [AI Seminar] CSD Distinguished Lecture 7/21/15 -Ariel Procaccia - Assistant Professor- Computer Science Department- Carnegie Mellon University Message-ID: CSD Community, There will be a CSD Distinguished Lecture July 21st, 2015 in NSH 3305 at 2pm. Ariel Procaccia will be giving a talk on AI and Economics. Talk title/abstract/bio are below for your review: "AI and Economics for a Healthier, Safer, and Fairer World" Abstract: I will explore the broad and exciting interaction between AI and economics, which spans the spectrum from deep theory to deployed applications in healthcare, physical security, and dispute resolution. Specifically, I will talk about market design, focusing on kidney exchange algorithms that overcome uncertainty; game theory, with an emphasis on the problem of learning to play security games; and fair division, where I will argue that computational thinking gives rise to new notions of fairness, and show how these ideas are integrated into Spliddit.org, a not-for-profit website that offers provably fair solutions to everyday problems. On the way I will make a special effort to highlight new challenges for AI research. Bio: Ariel Procaccia is an assistant professor in the Computer Science Department at Carnegie Mellon University. He usually works on problems at the interface of computer science and economics. His distinctions include the IJCAI Computers and Thought Award (2015), the Sloan Research Fellowship (2015), the NSF Early Career Development Award (2014), and the IFAAMAS Victor Lesser Distinguished Dissertation Award (2009). Samantha DiNardo Administrative Assistant Carnegie Mellon University Computer Science Department Gates-Hillman Center - GHC-9118 5000 Forbes Avenue Pittsburgh, PA 15213 412-268-7660 sdinardo at cs.cmu.edu Samantha DiNardo Administrative Assistant Carnegie Mellon University Computer Science Department Gates-Hillman Center - GHC-9118 5000 Forbes Avenue Pittsburgh, PA 15213 412-268-7660 sdinardo at cs.cmu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From sdinardo at cs.cmu.edu Sun Jul 19 20:18:04 2015 From: sdinardo at cs.cmu.edu (Samantha J Dinardo) Date: Sun, 19 Jul 2015 20:18:04 -0400 Subject: [AI Seminar] Reminder: CSD Distinguished Lecture Tuesday 7/21/15 -Ariel Procaccia - Assistant Professor- Computer Science Department- Carnegie Mellon University In-Reply-To: References: Message-ID: <5B5A26C2-3EAF-489B-AB16-86D925F49E29@cs.cmu.edu> CSD Community, There will be a CSD Distinguished Lecture at 2pm on Tuesday July 21st, 2015 in NSH 3305. Ariel Procaccia will be giving a talk on AI and Economics. Talk title/abstract/bio are below for your review: ?AI and Economics for a Healthier, Safer, and Fairer World? Abstract: I will explore the broad and exciting interaction between AI and economics, which spans the spectrum from deep theory to deployed applications in healthcare, physical security, and dispute resolution. Specifically, I will talk about market design, focusing on kidney exchange algorithms that overcome uncertainty; game theory, with an emphasis on the problem of learning to play security games; and fair division, where I will argue that computational thinking gives rise to new notions of fairness, and show how these ideas are integrated into Spliddit.org, a not-for-profit website that offers provably fair solutions to everyday problems. On the way I will make a special effort to highlight new challenges for AI research. Bio: Ariel Procaccia is an assistant professor in the Computer Science Department at Carnegie Mellon University. He usually works on problems at the interface of computer science and economics. His distinctions include the IJCAI Computers and Thought Award (2015), the Sloan Research Fellowship (2015), the NSF Early Career Development Award (2014), and the IFAAMAS Victor Lesser Distinguished Dissertation Award (2009). Samantha DiNardo Administrative Assistant Carnegie Mellon University Computer Science Department Gates-Hillman Center - GHC-9118 5000 Forbes Avenue Pittsburgh, PA 15213 412-268-7660 sdinardo at cs.cmu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From yano at cs.cmu.edu Tue Jul 21 13:55:32 2015 From: yano at cs.cmu.edu (Charlotte Yano) Date: Tue, 21 Jul 2015 13:55:32 -0400 Subject: [AI Seminar] Reminder: CSD Distinguished Lecture Tuesday 7/21/15 -Ariel Procaccia - Assistant Professor- Computer Science Department- Carnegie Mellon University In-Reply-To: <5B5A26C2-3EAF-489B-AB16-86D925F49E29@cs.cmu.edu> References: <5B5A26C2-3EAF-489B-AB16-86D925F49E29@cs.cmu.edu> Message-ID: Starting now, please join! -Charlotte From: Cs-friends [mailto:cs-friends-bounces at mailman.srv.cs.cmu.edu] On Behalf Of Samantha J Dinardo Sent: Sunday, July 19, 2015 8:18 PM To: cmacs_scs-cmu at mailman.srv.cs.cmu.edu; cs-friends at cs.cmu.edu; cs-students at cs.cmu.edu; pop-seminar at cs.cmu.edu; theory-announce at cs.cmu.edu; intelligence-seminar-announce at mailman.srv.cs.cmu.edu; cs-econ at cs.cmu.edu Subject: Reminder: CSD Distinguished Lecture Tuesday 7/21/15 -Ariel Procaccia - Assistant Professor- Computer Science Department- Carnegie Mellon University CSD Community, There will be a CSD Distinguished Lecture at 2pm on Tuesday July 21st, 2015 in NSH 3305. Ariel Procaccia will be giving a talk on AI and Economics. Talk title/abstract/bio are below for your review: "AI and Economics for a Healthier, Safer, and Fairer World" Abstract: I will explore the broad and exciting interaction between AI and economics, which spans the spectrum from deep theory to deployed applications in healthcare, physical security, and dispute resolution. Specifically, I will talk about market design, focusing on kidney exchange algorithms that overcome uncertainty; game theory, with an emphasis on the problem of learning to play security games; and fair division, where I will argue that computational thinking gives rise to new notions of fairness, and show how these ideas are integrated into Spliddit.org, a not-for-profit website that offers provably fair solutions to everyday problems. On the way I will make a special effort to highlight new challenges for AI research. Bio: Ariel Procaccia is an assistant professor in the Computer Science Department at Carnegie Mellon University. He usually works on problems at the interface of computer science and economics. His distinctions include the IJCAI Computers and Thought Award (2015), the Sloan Research Fellowship (2015), the NSF Early Career Development Award (2014), and the IFAAMAS Victor Lesser Distinguished Dissertation Award (2009). Samantha DiNardo Administrative Assistant Carnegie Mellon University Computer Science Department Gates-Hillman Center - GHC-9118 5000 Forbes Avenue Pittsburgh, PA 15213 412-268-7660 sdinardo at cs.cmu.edu -------------- next part -------------- An HTML attachment was scrubbed... URL: From awm at cs.cmu.edu Tue Nov 3 10:03:23 2015 From: awm at cs.cmu.edu (Andrew Moore) Date: Tue, 3 Nov 2015 10:03:23 -0500 Subject: [AI Seminar] Fwd: KLRN Virtual Classroom Screening In-Reply-To: <9697FFBACBCE9442B7CF9CA13F60A301023ADF399E@elmo> References: <9697FFBACBCE9442B7CF9CA13F60A301023ADF399E@elmo> Message-ID: Dear AI faculty and AI friends, I'm wondering if anyone would be interested in volunteering for a medium sized event with some middle school and high school kids tomorrow. The deal is one hour spent over chat (I think it's chat as opposed to live videoconference) with a total of 50-100 school kids from around the country, who will have just seen a nova show which focusses on the future world with AI and robotics and will have a bunch of questions. Please let me and Katrina (cc'd) know if you'd be interested in helping out at one of the two times below (warning, note they are central). I think it would be fun and I'd do it myself except I am enmeshed in meeting all of tomorrow. Andrew ---------- Forwarded message ---------- From: Katrina Kehoe Date: Mon, Nov 2, 2015 at 2:02 PM Subject: KLRN Virtual Classroom Screening To: "awm at cs.cmu.edu" Cc: Jennifer Herrera Hi Professor Moore, It was great to speak with you and I hope that someone from you university can join us on our two online screenings this week. Below is information to share: - This week?s program: NOVA Science Now: What?s The Next Big Thing Watch the full episode here: http://video.klrn.org/video/1801365037/ - There are two screenings on Wednesday, November 4 11AM CST https://ovee.itvs.org/screenings/jy3uo 3:30PM CST https://ovee.itvs.org/screenings/kgb78 - Our screening is about 1 hour long and that includes a question/answer session for the students. - We usually have between 50-80 students on with us each week---sometimes full classes join! - We would like at least 1 person from your university on each screening?feel free to have more than 1 person join us if you feel like it would be beneficial. All we need is a name, headshot, short bio, and e-mail address for each person who joins us from your university. - The students are usually VERY engaged and ask fun questions and the questions are usually typical middle and high school questions?entertaining and easy. Attached is a graphic that we use to promote this week?s class. Feel free to share to help to explain what our objective is. Thanks so much for your help! I look forward to hearing back from you or someone from your department. Feel free to call or e-mail me with questions. All the best, Katrina * Katrina Kehoe | VP Communications and Marketing KLRN-TV 210.208.8433 <210.208.8433>* *kkehoe at klrn.org * *klrn.org * -- Andrew Moore , Dean, School of Computer Science , Carnegie Mellon. Twitter feed -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image001.jpg Type: image/jpeg Size: 41661 bytes Desc: not available URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: KLRN Virtual Classroom SQUARED - Big_Nov4.jpg Type: image/jpeg Size: 435873 bytes Desc: not available URL: From KKehoe at klrn.org Tue Nov 3 10:23:04 2015 From: KKehoe at klrn.org (Katrina Kehoe) Date: Tue, 3 Nov 2015 15:23:04 +0000 Subject: [AI Seminar] KLRN Virtual Classroom Screening In-Reply-To: References: <9697FFBACBCE9442B7CF9CA13F60A301023ADF399E@elmo> Message-ID: <9697FFBACBCE9442B7CF9CA13F60A301023ADF5E71@elmo> Hi all. Yes, thank you Andrew for the introduction. It is actually a chat session (you are correct, no video conference (?so you can participate from home, a coffee shop, library, office or anywhere!). You will actually be watching the Nova Science Now episode and chatting at the same time. Below is a screen shot of a previous classroom so you can get an idea of what it is like. Your participation would make the experience amazing for the students. Please let me know if you are able to join us. Many thanks! Katrina [cid:image002.jpg at 01D11619.3DBAC9A0] From: awmcmu at gmail.com [mailto:awmcmu at gmail.com] On Behalf Of Andrew Moore Sent: Tuesday, November 03, 2015 9:03 AM To: ai-seminar-announce at cs.cmu.edu; Manuela Veloso; Catherine Copetas; Tuomas Sandholm Cc: Katrina Kehoe Subject: Fwd: KLRN Virtual Classroom Screening Dear AI faculty and AI friends, I'm wondering if anyone would be interested in volunteering for a medium sized event with some middle school and high school kids tomorrow. The deal is one hour spent over chat (I think it's chat as opposed to live videoconference) with a total of 50-100 school kids from around the country, who will have just seen a nova show which focusses on the future world with AI and robotics and will have a bunch of questions. Please let me and Katrina (cc'd) know if you'd be interested in helping out at one of the two times below (warning, note they are central). I think it would be fun and I'd do it myself except I am enmeshed in meeting all of tomorrow. Andrew ---------- Forwarded message ---------- From: Katrina Kehoe > Date: Mon, Nov 2, 2015 at 2:02 PM Subject: KLRN Virtual Classroom Screening To: "awm at cs.cmu.edu" > Cc: Jennifer Herrera > Hi Professor Moore, It was great to speak with you and I hope that someone from you university can join us on our two online screenings this week. Below is information to share: - This week?s program: NOVA Science Now: What?s The Next Big Thing Watch the full episode here: http://video.klrn.org/video/1801365037/ - There are two screenings on Wednesday, November 4 11AM CST https://ovee.itvs.org/screenings/jy3uo 3:30PM CST https://ovee.itvs.org/screenings/kgb78 - Our screening is about 1 hour long and that includes a question/answer session for the students. - We usually have between 50-80 students on with us each week---sometimes full classes join! - We would like at least 1 person from your university on each screening?feel free to have more than 1 person join us if you feel like it would be beneficial. All we need is a name, headshot, short bio, and e-mail address for each person who joins us from your university. - The students are usually VERY engaged and ask fun questions and the questions are usually typical middle and high school questions?entertaining and easy. Attached is a graphic that we use to promote this week?s class. Feel free to share to help to explain what our objective is. [cid:image003.jpg at 01D11619.3DBAC9A0] Thanks so much for your help! I look forward to hearing back from you or someone from your department. Feel free to call or e-mail me with questions. All the best, Katrina Katrina Kehoe | VP Communications and Marketing KLRN-TV 210.208.8433 kkehoe at klrn.org klrn.org -- Andrew Moore, Dean, School of Computer Science, Carnegie Mellon. Twitter feed -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image002.jpg Type: image/jpeg Size: 68526 bytes Desc: image002.jpg URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: image003.jpg Type: image/jpeg Size: 41661 bytes Desc: image003.jpg URL: From copetas at cs.cmu.edu Tue Dec 1 16:41:06 2015 From: copetas at cs.cmu.edu (Catherine Copetas) Date: Tue, 1 Dec 2015 16:41:06 -0500 Subject: [AI Seminar] Heinz Faculty Research Seminar: Dec 7 - Patrick Bajari, VP and Chief Economist, Amazon Message-ID: <565E13F2.6010607@cs.cmu.edu> Of special interest.... Patrick Bajari is Vice President and Chief Economist for Amazon. He will be the speaker at the Heinz Faculty Research Seminar on Monday, *December 7th, 2015 at 12:00 noon in room 1000, Hamburg Hall*. Below you will find the details for his talk. Please distribute this announcement to appropriate lists in your departments. Title: *Demand Estimation in a High Dimensional Data and an Application to Amazon?s Book Marketplace * Abstract: In this paper, we document some basic facts about Amazon?s e-commerce platform using several unique datasets. There are both sizeable applied theory and empirical literatures which address e-commerce. We contribute to the empirical literature by examining datasets from Amazon that contain the entire population of product sales, product offers and sellers from multiple countries and for multiple product lines. We also include important control variables that have been omitted from previous research. We begin by examining a dataset of all physical books and e-books that were buyable on Amazon as of November 2014. The selection available to consumers is large - there were 44 million unique physical book titles available for purchase in the US and over 33 million unique physical book titles in the EU in November 2014. Approximately two million e-books are self-published accounting for two thirds of total selection. Next, following the method in Brynjolfsson, Hu and Smith (2003) we estimate the compensating variation for book selection available on Amazon?s book marketplace. We apply a novel estimation procedure to obtain elasticities required for this calculation which incorporates a mixture of traditional econometric approaches and machine learning approaches. We estimate an annualized 2014 benefit to consumers of $2.2 billion in the US and $2.1 billion in the EU. We continue by documenting some basic facts about the supply side of the platform. Specifically, we summarize the number of offers by seller in select product categories, summarize the number of distinct 3rd party sellers listing products on Amazon, and provide estimates for the number of employees at 3rd party sellers. An average book in our sample has 20 competing offers. Consistent with a large number of multiple offers we find that the median residual elasticity ranges from -7 to -32 for several product lines in multiple countries. There were over 1.1 million merchants with offers on books on November 2014 and hundreds of thousands of merchants offering books in the EU. Finally, we estimate the number of full time employees directly attributable to all sellers in all product lines. We estimate 414,550 employees at 3rd part sellers in the US and 233,612 employees at 3rd party sellers in the EU in 2014. Bio: Pat Bajari is Vice President and Chief Economist for Amazon. He earned his doctorate in economics in 1997 from the University of Minnesota where he specialized in applied econometrics and empirical industrial organization. He was a member of the economics faculty at Harvard, Stanford, Duke and Minnesota and is a research fellow at the NBER. Pat worked for the Boston Consulting Group from 1999 to 2010 in pricing, providing applied econometric work to guide decision making in a variety of industries. He also worked as a consultant to the research departments for the Federal Reserve Banks of San Francisco and Minneapolis and served as an antitrust expert witness. Pat joined Amazon in 2010 where he reports to Jeff Wilke, the Senior Vice President for the North American Consumer Platform. He has used economic and econometric tools to build automated systems for pricing, forecasting, ordering, marketing spend in addition to providing applied econometric analysis on a host of other business problems. -------------- next part -------------- An HTML attachment was scrubbed... URL: From evitercik at cs.cmu.edu Tue Dec 22 12:54:44 2015 From: evitercik at cs.cmu.edu (Ellen Vitercik) Date: Tue, 22 Dec 2015 12:54:44 -0500 Subject: [AI Seminar] AI lunch call for talks Message-ID: Hi everyone! Ariel and I will be organizing AI lunch beginning this spring semester. Lunches will be held every Tuesday from noon to 1 in NSH 3305, unless we announce otherwise. We hope to have outside speakers come about once per month and we encourage students and post-docs to give talks during the remaining weeks. AI lunch will be a great venue for you to present your work to a broad AI audience, so we hope that you will help us ensure a successful start to this new tradition by volunteering to give a talk. If you are interested in giving a 50 minute talk, please email me at evitercik at cs.cmu.edu with your scheduling constraints. Best, Ellen and Ariel -------------- next part -------------- An HTML attachment was scrubbed... URL: