|  |
演讲简介
Artificial intelligence (AI) models trained on published scientific findings have been used to invent valuable materials and targeted therapies, but they typically ignore the human scientists who continually alter the landscape of discovery. Here we show that incorporating the distribution of human expertise by training unsupervised models on simulated inferences cognitively accessible to experts dramatically improves (up to 400%) AI prediction of future discoveries beyond those focused on research content alone, especially when relevant literature is sparse. These models succeed by predicting human predictions and the scientists who will make them. By tuning human-aware AI to avoid the crowd, we can generate scientifically promising "alien" hypotheses unlikely to be imagined or pursued without intervention until the distant future, which hold promise to punctuate scientific advance beyond questions currently pursued. Accelerating human discovery or probing its blind spots, human-aware AI enables us to move toward and beyond the contemporary scientific frontier.
关于讲者
James Evans’ research focuses on the collective system of thinking and knowing, ranging from the distribution of attention and intuition, the origin of ideas and shared habits of reasoning to processes of agreement (and dispute), accumulation of certainty (and doubt), and the texture—novelty, ambiguity, topology—of understanding. He is especially interested in innovation—how new ideas and practices emerge—and the role that social and technical institutions (e.g., the Internet, markets, collaborations) play in collective cognition and discovery. Much of James Evan’s work has focused on areas of modern science and technology, but he is also interested in other domains of knowledge—news, law, religion, gossip, hunches, machine and historical modes of thinking and knowing. support the creation of novel observatories for human understanding and action through crowd sourcing, information extraction from text and images, and the use of distributed sensors (e.g., RFID tags, cell phones). James Evans uses machine learning, generative modeling, social and semantic network representations to explore knowledge processes, scale up interpretive and field- methods, and create alternatives to current discovery regimes. His research has been supported by the National Science Foundation, the National Institutes of Health, the Air Force office of Science Research, and many philanthropic sources, and has been published in Nature, Science, Proceedings of the National Academy of Science, American Journal of Sociology, American Sociological Review, Social Studies of Science, Research Policy, Critical Theory, Administrative Science Quarterly, and other outlets. His work has been featured in the Economist, Atlantic Monthly, Wired, NPR, BBC, El País, CNN, Le Monde, and many other outlets.
At Chicago, James Evans is the Director of Knowledge Lab, which has collaborative, granting and employment opportunities, as well as ongoing seminars. He also founded and now direct on the Computational Social Science program at Chicago, and sponsor an associated Computational Information of Participants Social Science workshop. James Evans teaches courses in augmented intelligence, the history of modern science, science studies, computational content analysis, and Internet and Society. Before Chicago, James Evans received my doctorate in sociology from Stanford University, served as a research associate in the Negotiation, Organizations, and Markets group at Harvard Business School, started a private high school focused on project-based arts education, and completed a B. A. in Anthropology at Brigham Young University.
