Hello TUX!
A reminder that tomorrow we have a Member Presentation by Prof. Ron Baecker.
We look forward to seeing you there!
Ali, Daniel, and Tovi
TUX Member Presentation: Prof. Ron Baecker
March 27, 2018. DGP Lab, Department of Computer Science at U of T @ 40 St. George St. Room 5166
Lunch reception begins at 12:30 pm. Presentation begins at 1:00 pm.
AI Deployments Accelerate Without Sufficient Intelligence: Opportunities for HCI Research
After 35 years of ups and downs, AI finally achieved in the last two decades triumphs over the world's best humans in chess, Jeopardy, Go, and poker. Accelerating advances in deep leaning technology now bring actual or promised deployments in speech and face recognition, judgments of human potential, medical image processing, driverless cars, and automated warfare. But are these systems truly intelligent? Replacing simplistic definitions of intelligence with the more nuanced descriptions of Sternberg and Gardner suggests that the answer is "no". Thinking about what we should expect of intelligent agents, we must acknowledge the lack of algorithms that can explain the logic behind their actions so that we can understand their behaviour. This is required so that we can trust them, delegate responsibility for actions and accountability for errors, and expect their decisions that are just. Removing these limitations will require a healthy dose of HCI research and user experience innovation. My goal with this talk is to encourage audience members to work on these issues.
Note: Ideas in this lecture are based in part on Chapter 11 of the forthcoming text: Computers and Society: Modern Perspectives, by Ronald M. Baecker, Oxford University Press, 2019.
Bio
Ronald Baecker is Director of the Technologies for Aging Gracefully Laboratory (TAGlab), Professor of Computer Science, and Bell Universities Laboratories Chair in Human-Computer Interaction.
The focus of TAGlab activities is R&D in support of aging throughout the life course including cognition, communication, and social interaction. Collaborators include individuals from Baycrest, Columbia Medical School, Sunnybrook Health Sciences Centre, and Toronto Rehabilitation Institute.
He is also Affiliate Scientist with the Kunin-Lunenfeld Applied Research Unit of Baycrest (formerly, Baycrest Centre for Geriatric Care), Adjunct Scientist with Toronto Rehabilitation Institute, and founder of the Knowledge Media Design Institute at the University of Toronto. He has been named one of the 60 Pioneers of Computer Graphics by ACM SIGGRAPH, has been elected to the CHI Academy by ACM SIGCHI, and has been given the Canadian Human Computer Communications Society Achievement Award in May 2005. His B.Sc., M.Sc., and Ph.D. are from the Massachusetts Institute of Technology.
Professor Baecker is an active researcher, lecturer, and consultant on human-computer interaction and user interface design, user support, software visualization, multimedia, computer-supported cooperative work and learning, the Internet, entrepreneurship and strategic planning in the software industry, and the role of information technology in business. He has published over 175 papers and articles on topics in these areas. He is also the author or co-author of two published videotapes and of four books:
. "Reading in Human-Computer Interaction: A Multidisciplinary Approach",
. "Human Factors in Typography for More Readable Programs",
. "Readings in Groupware and Computer-Supported Cooperative Work: Facilitating Human-Human Collabortation", and
. "Reading in Human-Computer Interaction: Towards the Year 2000".
He is the co-holder of 2 patents. Professor Baecker was the founder, CEO, and Chairman of HCR Corporation, a Toronto-based UNIX contract R&D and technology development and marketing firm, sold in 1990 to a U.S. competitor. He was also the founder of Expresto Software Corp, a firm specializing in structured visual communication explaining software and other complex technology. Expresto Software was sold in 2002 to Caseware International. Another entrepreneurial venture was a virtual non-profit foundation within the University of Toronto to distribute and support the open source ePresence Interactive Media rich media webcasting and archiving system, which then led to the formation of a start-up delivering ePresence products, services, and solutions, recently sold to Desire2Learn. Most recently, he was instrumental in the founding of MyVoice.
OUR SPONSORS:
TUX is made possible by the support of our sponsors, Steven Sanders, Autodesk,
University of Toronto Department of Computer Science, and MaRS.
About MaRS: MaRS is the one of the world's largest urban innovation hubs-a place for collaboration, creativity and entrepreneurship. Located in the heart of Toronto's research district, MaRS provides the space, training, talent and networks required to commercialize important discoveries and launch and grow Canadian startups.
cid:image005.png@01D357B6.5B82A7B0
cid:image008.png@01D10810.C2D01210
cid:image008.png@01D357B6.5B82A7B0
Hello TUX!
A reminder that tomorrow we have a Member Presentation by Prof. Joseph Jay Williams.
We look forward to seeing you there!
Ali, Daniel, and Tovi
TUX Member Presentation: Prof. Joseph Jay Williams
*September 25, 2018. Autodesk MaRS IDEaS Lab, 661 University Ave., Ground Floor*
**
Lunch reception begins at 12:30 pm. Presentation begins at 1:00 pm.
*Combining Active Learning & Human Computation for A/B Experimentation: Perpetually Enhancing and Personalizing User Interfaces
How can we transform the everyday technology people use into intelligent, self-improving systems? I consider how to dynamically enhance user interfaces by using randomized A/B experiments to integrate Active Learning algorithms with Human Computation. Multiple components of a user interface (e.g. explanations, messages) can be crowdsourced from users, and then compared in real-world A/B experiments, bringing human intelligence into the loop of system improvement. Active Learning algorithms (e.g. multi-armed bandits) can then analyze data from A/B experiments in order to dynamically provide more effective A or B conditions to future users. Active Learning can also lead to personalization, by facing the more substantive exploration-exploitation tradeoff of discovering whether some conditions work better for certain subgroups of user profiles (in addition to discovering what works well on average).
I present an example system, which crowdsourced explanations for how to solve math problems from students and teachers, simultaneously conducting an A/B experiment to identify which explanations other students rated as being helpful. Modeling this as a multi-armed bandit where the arms were constantly increasing (every time a new explanation was crowdsourced) we used Thompson Sampling to do real-time analysis data from the experiment, providing higher rated explanations to future students (LAS 2016, CHI 2018). This generated explanations that helped learning as much as those of a real instructor. Future work aims to discover how to personalize explanations in real-time, by discovering which conditions work for different subgroups of user profiles (such as whether simple vs complex explanations are better for students with different levels of prior knowledge or verbal fluency).
Future collaborative work with statistics and machine learning researchers provides a testbed for a wide range of active learning algorithms to do real-time adaptation of A/B experiments, and integrate with different crowdsourcing workflows. Dynamic A/B experiments can be used to enhance and personalize a broad range of user-facing systems. Examples include modifying websites, tailoring email campaigns, enhancing lessons in online courses, getting people to exercise by personalizing motivational messages in mobile apps, and discovering which interventions reduce stress and improve mental health.
*Bio*Joseph Jay Williams is an Assistant Professor in Computer Science at the University of Toronto. He was previously an Assistant Professor at the National University of Singapore’s School of Computing in the department of Information Systems & Analytics, a Research Fellow at Harvard’s Office of the Vice Provost for Advances in Learning, and a member of the Intelligent Interactive Systems Group in Computer Science. He completed a postdoc at Stanford University in Summer 2014, working with the Office of the Vice Provost for Online Learning and the Open Learning Initiative. He received his PhD from UC Berkeley in Computational Cognitive Science, where he applied Bayesian statistics and machine learning to model how people learn and reason. He received his B.Sc. from University of Toronto in Cognitive Science, Artificial Intelligence and Mathematics, and is originally from Trinidad and Tobago. More information about his research and papers is at www.josephjaywilliams.com.
cid:image009.png@01D357B6.5B82A7B0
*OUR SPONSORS:*
cid:image010.png@01D357B6.5B82A7B0
*TUX is made possible by the support of our sponsors, Steven Sanders, Autodesk, *
*University of Toronto Department of Computer Science, and MaRS*.
*/About MaRS:/*/MaRS is the one of the world’s largest urban innovation hubs—a place for collaboration, creativity and entrepreneurship. Located in the heart of Toronto’s research district, MaRS provides the space, training, talent and networks required to commercialize important discoveries and launch and grow Canadian startups./
tux-announce mailing list tux-announce@dgp.toronto.edu https://www.dgp.toronto.edu/cgi-bin/mailman/listinfo/tux-announce
-
Sharon Tan -
Xuan Dam