Technology has always been in the service of augmenting human capabilities. AI is our way of doing for mental work what the wheel, the plow, and the steam engine did for physical labor. Yet machine learning has the capability of going its own way, whether it’s Facebook bots inventing their own language or swarms of drones making collective decisions that can’t be attributed to any one node. What is the current state of the art, and how can we best harness its growing capabilities?
Smart everything. Data analysis. Human understanding. Human learning (e.g., “Algorithms to Live By”). Discovery. Programmable matter. Natural language processing. Visual recognition. TPUs and machine learning APIs. New networks and surveillance capitalism. Prediction. Neural Silicon Hybrids. Robot self-awareness. Complexity. AI and the future of work.
Erik Andrejko, Chief Technology Officer, Wellio — Intelligent assistants
Elias Bareinboim, Purdue— Causal inference and data-fusion
Noam Brown, Carnegie Mellon— Libratus: the winning poker bot
David Gunning, DARPA — The Explainable AI project
Guy Hoffman, Cornell — A handmade approach to social robotics
Jeff Jonas, Senzing — Big data and the GDPR
Ben Kuipers, University of Michigan — How can we trust a robot?
Doug Lenat, Cyc — Compensating for cognitive brittleness
Gary Marcus, NYU — Why AI is Harder Than You Think
David Prior, Xuvasi — Reorganizing business processes using applied intelligence
Ken Stanley, Uber AI Labs and University of Central Florida —Neuroevolution and the quest for AI
Simon Tong, FiveAI — Autonomous vehicles without maps
Bo Zhu, Harvard —Image reconstruction from MRIs to radio astronomy: a unified framework
Anthony Zador, M.D., Cold Spring Harbor Laboratory —Building better brain-like algorithms |
