AI Futures: Report from the One Hundred Year Study on Artificial Intelligence (AI100)
The One Hundred Year Study on Artificial Intelligence, or AI100, has a unique mission: launch a study every five years, over the course of a century, to better track and anticipate how artificial intelligence ripples through society, shaping every facet of how people work, live, and play. The first study panel report, released in 2016, was well-received by researchers, educators, practitioners, and policy-makers. The second study panel report, expected in late 2021, is now underway. It will be based, in part, on two study-workshops commissioned by the AI100 standing committee, one entitled “Coding Caring” and the other “Prediction in Practice”. This IJCAI roundtable will bring together members of the standing committee, the ongoing study panel, and the two workshops for a discussion about the themes of those workshops, the current study panel’s efforts, and the overall trajectory of AI100.
Peter Stone and Mary Gray
January 13th, 12:45-13:45 JST
AI in FinTech: Challenges and future
AI in finance has been a lasting and increasingly vigorous area. It goes beyond the applications of AI in finance services. The era of new-generation AI inflames and embraces unprecedented financial innovations that facilitate, diversify, and transform our daily life, society, and economy. While enjoying and excited about numerous new AI applications in finance, one may be more curious about:
- What makes AI unrivaled in innovating FinTech?
- What financial challenges demand AI?
- What are new challenges brought by AI-empowered finance?
- What will next-generation AI-enabled FinTech look like? and
- Will AI bring more opportunities or challenges to future finance and economy?
Answering these big questions demand deep thinking, insight, knowledge, and experience in the interdisciplinary and cross-domain research, innovation, and practices. Four outstanding leaders from both academia and industry will share their unique insights and impactful experience on the above issues and more. You are welcome to attend this prestigious panel and join the webinar discussions.
- Dr Usama Fayyad, CEO, Open Insights, USA
- Dr Amy Shi-Nash, Global Head of Analytics and Data Science at HSBC, UK
- Prof Michael Wellman, Professor, University of Michigan, USA
- Prof Qiang Yang, Chief AI Officer, WeBank/Chair Professor at HKUST, China
January 14th, 8:00-9:00 JST
Diversity in AI
AI has already a large impact on the life of every person and this impact will grow with digitalization and automation in the future. Thus, any development and application of AI must take into account the diversity of humans, making AI research and AI products accessible for everybody. We need to be aware that there are under-represented communities within and without of AI. Is this acceptable considering today’s global impact?
In the panel and following table discussions we plan to open up questions such as:
- What does “under-represented” actually entail?
- What, if increasing awareness is not enough?
- Diversity in Research – are current best practices really best?
- AI for diverse users – how to integrate accessibility considerations systematically and always?
Neil Yorke-Smith and Franziska Klügl
- Bo An, Nanyang Technological University, Singapore
- Tanuja Ganu, Microsoft Research, India
- Xin Geng, Southeast University, China
January 12th, 12:45-13:45 JST
Computational Sustainability and Human Well-being
Computational Sustainability is an interdisciplinary field that applies techniques from a broad set of disciplines like computer science, information science, operations research, applied mathematics, and statistics to solve global challenges for balancing environmental, economic, and societal needs for sustainable development.
The main aim is to develop computational and mathematical models and methods for decision making concerning the management and allocation of resources in order to help solve some of the most challenging problems related to sustainability.
There are a number of open questions in this field.
The scale and complexity of computational sustainability problems are extremely challenging. Problems are large-scale with dynamic aspects and uncertainties regarding the input data. Hence exact algorithms are rarely applicable and instead heuristics and machine learning methods are wideley used. Should we nevertheless be more concerned with quality guarantees for the computed solutions? And if so, what standards for experiments or case studies would be desirable?
How could we strengthen the incentives to trust and use computational sustainability solutions in the real-world, e.g. to inform policies or planning and decision making processes? Should there be an increased focus on creating ready-to-use implementations and tools?
What courses should ideally be taught to students to make them well-prepared for working on inherently multi-disciplinary computational sustainability problems?
Michela Milano and Sabine Storandt
- Carla Gomes
- Pascal Van Hentenryck
- Barry O’Sullivan
January 12th, 19:00-20:00 JST
AI approaches for Covid-19
COVID-19 (or coronavirus) is the greatest public health crisis that the world has experienced in the last century. Tackling it effectively requires the collective will of experts from a variety of disciplines. From a computer science perspective, Artificial Intelligence researchers have traditionally been at the forefront of responding to rapidly emerging societal needs. The panel will engage attendees in meaningful discussions around the different ways in which AI/ML research can help in tackling the COVID-19 pandemic.
Clearly, AI researchers have already made rapid strides in this space. For example, a lot of efforts have been made by AI researchers in developing agent-based models for simulating the transmission of COVID-19. In addition, AI researchers have proposed Markovian models for generated optimal sequential intervention strategies for effective imposition and release of lockdown protocols. At the same time, we have not yet seen the enormous potential of AI research being leveraged to design decision support systems (e.g., in the allocation of limited healthcare resources such as testing kits) which can assist epidemiologists and policy-makers in their fight against this pandemic.
While there exist many different areas in which AI can help in the fight against this pandemic, this panel will bring together world-reknowned AI and epidemiology experts to illustrate five different strands of AI research that can help in the fight against COVID-19. In particular, the panelists will be focusing on (i) AI for transmission modeling; (ii) AI for COVID-19 forecasting; (iii) AI for optimal healthcare resource utilization; (iv) AI (or NLP) on CORD-19 dataset; and (v) AI for Therapeutics and Vaccine Design. The panel’s duration is one hour. At the beginning, each panelist will be making a short opening statement to set the context. Followed by that, there would be 30-40 minutes of Q&A with the audience, so the panel is meant to be highly interdisciplinary.
Amulya Yadav, PNC Technologies Career Development Assistant Professor, Pennsylvania State University
- Eric Horvitz, Technical Fellow & Chief Scientific Officer, Microsoft
- Oren Etzioni, Chief Executive Officer at Allen Institute for AI (AI2)
- Milind Tambe, Gordon McKay Professor of Computer Science, Harvard University & Director “AI for Social Good”, Google Research India
- Roni Rosenfeld, Professor & Head, Machine Learning Department, School of Computer Science, Carnegie Mellon University
- Maimuna Majumdar, Faculty at Computational Health Informatics Program, Boston Children’s Hospital & Harvard University
January 15th, 8:00-9:00 JST