AAAI-15 Invited Speakers

AAAI-15 Invited Speakers

AAAI-15 / IAAI-15 will feature the following series of distinguished speakers and panelists:

Tuesday, January 27
9:00 – 9:50 AM

• AAAI-15 Invited Talk: Drew Bagnell
• AAAI-15/IAAI-15 Joint Invited Talk: Oren Etzioni

5:45 PM – 7:15 PM

• AAAI-15 Shakey Celebration: Ed Feigenbaum, Peter Hart, and Nils Nilsson et al.

Wednesday, January 28
1:40 – 2:30 AM

• AAAI-15 Invited Talk: Geoffrey Hinton
• IAAI-15 Invited Talk: Rayid Ghani

Thursday, January 29
9:00 – 9:50 AM

• AAAI-15 Invited Talk: Meinolf Sellmann

5:45 - 6:45 PM

• AAAI-15 Debate on Autonomous Weapons: Ron Arkin, Stephen Goose

Friday, January 30
9:00 – 9:50 AM

• AAAI-15 Invited Talk: Lise Getoor
• AAAI-15 Invited Talk: Michael Bowling

AAAI-15 Invited Talk
Artificial Intelligence, Machine Learning and Robotics: Interplay and Interaction

Drew Bagnell (Carnegie Mellon University)

My talk will focus on theoretical and algorithmic ideas in machine learning and AI, and their origin in problems of robotics. Much of my talk will focus on no-regret online learning methods in machine learning and the critical role of interaction for learning in robotics.

I will highlight the tremendous impact robotics has had in identifying key learning problems and suggesting algorithmic techniques; additionally, I'll consider the remarkable tools that have been developed within AI and learning to address hard robotics problems. I'll discuss a variety of machine learning techniques of increasing sophistication from the most familiar classification problems, to structured prediction, and to imitation learning. I will also address how to make reinforcement learning and learning control practical in robotics. Throughout, we will look at case studies in learning dexterous manipulation, activity forecasting of drivers and pedestrians, and imitation learning of robotic locomotion and rough-terrain navigation. These case studies highlight key challenges in applying AI and learning algorithms in practical settings.

J. A. (Drew) Bagnell is an associate professor at Carnegie Mellon University's Robotics Institute, National Robotics Engineering Center and Machine Learning Department. His interests in artificial intelligence range from algorithmic and basic theoretical development to delivering fielded learning-based systems. Bagnell directs the Learning, AI, and Robotics Laboratory (LAIRLab) within the Robotics Institute. He also serves as the director of the Robotics Institute Summer Scholars program, a summer research experience in robotics for undergraduates throughout the world. Bagnell and his groups have received research awards in both the robotics and machine learning communities including at the International Conference on Machine Learning, Robotics Science and Systems, and International Conference on Robotics and Automation. His current projects focus on machine learning for dexterous manipulation, decision making under uncertainty, ground and aerial vehicle control, robot perception and computer vision. Prior to joining the faculty, Bagnell received his doctorate as an NSF Graduate Fellow at Carnegie Mellon in 2004 and completed undergraduate studies with highest honors in electrical engineering at the University of Florida.

IAAI-15 /AAAI-15 Joint Invited Talk
You Can't Play 20 Questions with Nature and Win

Oren Etzioni (Allen Institute for Artificial Intelligence)

The machine learning paradigm, and deep learning methods in particular, have achieved phenomenal results in recent years. We need to leverage and extend these methods to address grand AI challenges such as the automated acquisition of common-sense knowledge, and machine reading of text. My talk will describe the ambitious research program at the Allen Institute of AI (AI2), which aims to address these challenges in collaboration with the AI community. I will describe our key projects: Aristo — which learns from textbooks and reasons over learned knowledge to answer standardized test questions, and Semantic Scholar — which aims to utilize AI methods to revolutionize the search for academic papers.

Oren Etzioni is CEO of the Allen Institute for Artificial Intelligence, and a professor at the University of Washington's Computer Science dDepartment. He has received the Robert Engelmore Memorial Award (2007), the IJCAI Distinguished Paper Award (2005), AAAI Fellow (2003), and a National Young Investigator Award (1993). He was also the founder of several companies including Farecast (sold to Microsoft in 2008) and Decide (sold to eBay in 2013), and is the author of more than 100 technical papers that have garnered roughly 22,000 citations. Etzioni received his Ph.D. from Carnegie Mellon University in 1991, and his BA from Harvard in 1986.

AAAI-15 Panel
The Shakey Celebration

Panelists: Ed Feigenbaum, Peter Hart, and Nils Nilsson

The Shakey Celebration will include a panel, along with other highlights of this historic project.

AAAI-15 Invited Talk
Deep Learning

Geoffrey Hinton (University of Toronto and Google Inc.)

I will give a brief history of deep learning explaining what it is, what kinds of task it should be good for and why it was largely abandoned in the 1990's. I will then describe how ideas from statistical physics were used to make deep learning work much better on small datasets. Finally I will describe how deep learning is now used by Google for speech recognition and object recognition and how it may soon be used for machine translation.

Geoffrey Hinton received his PhD in artificial intelligence from Edinburgh in 1978. He spent five years at Carnegie-Mellon. In 1987 he moved to the Department of Computer Science at the University of Toronto. Since 2013, he has been splitting his time between the University of Toronto and Google. He was one of the researchers who introduced the back-propagation algorithm. His other contributions to neural network research include Boltzmann machines, distributed representations, time-delay neural nets, mixtures of experts, variational learning, products of experts and deep belief nets. His students used deep learning to change the way in which speech recognition and object recognition are done.

IAAI-15 Invited Talk
Data Science for Social Good: Using Your Powers To Make a Social Impact!

Rayid Ghani (University of Chicago)

The past few years have seen an increasing demand for machine learning/data mining/data science powers. That's wonderful for us data scientists but wouldn't the world be so much better if we also used our computational and analytical powers for social good? In this talk, I'll give examples from work going on around the world including from the summer fellowship program we started at University of Chicago on Data Science for Social Good to show that there are a lot of important social problems in the world that could use our help — from helping students graduate high school to helping disaster victims to improving health.

Rayid Ghani is the director of the Center for Data Science and Public Policy at the University of Chicago. He is also the cofounder of Edgeflip, an analytics and social media startup that is focused on helping nonprofits and social causes. Previously, Ghani was the chief scientist for the Obama 2012 election campaign focusing on analytics, data, and technology.

Ghani is focused on education, research, and developing approaches to use data and analytics for social causes, both with Edgeflip and the University of Chicago.He created and runs the Eric and Wendy Schmidt Data Science for Social Good Summer Fellowship, which brings together aspiring data scientists to work on data science projects in partnership with governments and nonprofits.

AAAI-15 Invited Talk
Intelligent Decisions

Meinolf Sellmann (IBM Thomas J. Watson Research Center)

At its best intelligence creates aesthetic and beauty, yet from a utilitarian perspective intelligence primarily serves the purpose of making better decisions. Today's prescriptive decision support systems are most effective when applied to specific recurring operational problems. Recent advances in AI technology have revived the vision of commercially viable cooperative strategic decision support systems. These cognitive systems integrate information retrieval, knowledge representation, interactive modelling, as well as social and self-learning capabilities with logic reasoning and probabilistic decision making under uncertainty. I provide a snapshot of the current technology status by showcasing several projects that ultimately aim at intelligent human-in-the-loop decision making.

Meinolf Sellmann is manager of the AI for optimization group in the cognitive computing department at IBM Watson Research. He received his doctorate degree from Paderborn University from where he went on to Cornell University as postdoctoral associate and Brown University as assistant professor. Sellmann has published over 60 articles in international conferences and journals with central contributions in symmetry breaking, global constraints, the integration of mathematical and constraint programming, search, autonomous algorithm configuration, and algorithm portfolios. He served as conference chair of CP 2007, PC chair of CPAIOR 2013, and associate editor of the Informs Journal on Computing. He received an NSF Early Career Award in 2007 and IBM Outstanding Technical Innovation Awards in 2013 and 2014. Based on their meta-algorithmic research, for four years in a row Sellmann and his team won at international SAT and MaxSAT solver competitions, among others for the overall most efficient parallel SAT Solver in 2013, and three first places at the 2014 MaxSAT Evaluation.

AAAI-15 Debate
Debate on Autonomous Weapons

Participants: Ron Arkin (Georgia Institute of Technology) and Stephen Goose (Human Rights Watch)
Moderator: Thomas G. Dietterich (Oregon State University)

Ronald C. Arkin is a regents' professor, director of the Mobile Robot Laboratory, and associate dean for Research in the College of Computing at Georgia Tech. He served as STINT visiting professor at KTH in Stockholm, sabbatical chair at the Sony IDL in Tokyo, and the Robotics and AI Group at LAAS/CNRS in Toulouse. His research interests include behavior-based reactive control and action-oriented perception for mobile robots and unmanned aerial vehicles, hybrid deliberative/reactive software architectures, robot survivability, multiagent robotic systems, biorobotics, human-robot interaction, robot ethics, and learning in autonomous systems. His latest book is Governing Lethal Behavior in Autonomous Robots. Arkin serves on numerous journal editorial boards and is the series editor for The MIT Press book series Intelligent Robotics and Autonomous Agents. He has provided expert testimony to the United Nations, the International Committee of the Red Cross, the Pentagon and others on Autonomous Systems Technology. Arkin served on the Board of Governors of the IEEE Society on Social Implications of Technology, the IEEE RAS AdCom, and is founding cochair of IEEE RAS TC on Robot Ethics. He is a distinguished lecturer for the IEEE Society on social implications of technology and a Fellow of the IEEE.

Stephen Goose, is executive director of Human Rights Watch's Arms Division, was instrumental in bringing about the 2008 convention banning cluster munitions, the 1997 treaty banning antipersonnel mines, the 1995 protocol banning blinding lasers, and the 2003 protocol requiring clean-up of explosive remnants of war. He and Human Rights Watch cofounded the International Campaign to Ban Landmines (ICBL), which received the 1997 Nobel Peace Prize. Goose created the ICBL's Landmine Monitor initiative, the first time that non-governmental organizations around the world have worked together in a sustained and coordinated way to monitor compliance with an international disarmament or humanitarian law treaty. In 2013, he and Human Rights Watch cofounded the Campaign to Stop Killer Robots. Before joining Human Rights Watch in 1993, Goose was a US congressional staffer and a researcher at the Center for Defense Information. He has a master's degree in International Relations from the Johns Hopkins School of Advanced International Studies and a BA in History from Vanderbilt University.

Moderator Thomas G. Dietterich is the president of the Association for the Advancement of Artificial Intelligence.

AAAI-15 Invited Talk
Using Statistics and Semantics to Solve Big (Graph) Data Problems

Lise Getoor (University of California, Santa Cruz)

Big data problems benefit from modeling both structure and uncertainty, so there is a growing need for tools to develop large, complex probabilistic models. These tools should combine high-level knowledge representation with general purpose, scalable algorithms for learning and inference. In this talk, I will survey some of the recent work from the statistical relational learning community on learning and inference in richly-structured, multi-relational network data. I will highlight both important developments and opportunities in which ideas from AI can have great impact on upcoming challenges within the machine learning, data science and data mining communities.

Lise Getoor is a professor in the Computer Science Department at the University of California, Santa Cruz. Her research areas include machine learning and reasoning under uncertainty, with an emphasis on graph and network data. She is a AAAI Fellow, serves on the Computing Research Association and International Machine Learning Society boards, was cochair of ICML 2011, is a recipient of an NSF Career Award and eight best paper and best student paper awards. She received her PhD from Stanford University, her MS from the University of California, Berkeley, and her BS from the University of California, Santa Barbara, and was a professor at the University of Maryland, College Park from 2001–2013.

AAAI-15 Invited Talk
von Neumann's Dream

Michael Bowling (University of Alberta)

Chess has long served as the measure of progress for artificial intelligence. However, at the very beginning of computing and artificial intelligence, John von Neumann dreamt of a different game: "Real life is not like [chess]. Real life consists of bluffing, of little tactics of deception, of asking yourself what is the other man going to think I mean to do. And that is what games are about in my theory." The game von Neumann hinted at is poker, and it played a foundational role in his formalization of game theory. Shortly after launching the field of game theory, he practically abandoned his new discipline to focus on the budding field of computing. He saw computers as the way to make his mathematics workable. Now, over 70 years later with both significant advances in computing and game theoretic algorithms, von Neumann's dream is now a reality. Heads-up limit Texas hold'em poker, the smallest variant of poker played by humans, is essentially solved. In this talk, I will discuss how we accomplished this landmark result, along with the substantial scientific advances in our failed attempts along the way.

Michael Bowling is a professor of computing science at the University of Alberta. His research focuses on artificial intelligence, machine learning, and game theory; and he is particularly fascinated by the problem of how computers can learn to play games through experience. Bowling is the leader of the Computer Poker Research Group, which has built some of the strongest poker playing programs in the world. In 2008, one of these programs, Polaris, defeated a team of top professional poker players in two-player, limit Texas Hold'em, becoming the first program to defeat poker pros in a meaningful competition. His research has been featured on the television programs Scientific American Frontiers, National Geographic Today, and Discovery Channel Canada; in the New York Times and Wired; and twice in exhibits at the Smithsonian Museums in Washington, D.C.

Links