Ken Forbus, Professor of Electrical Engineering and Computer Science, and his team at Northwestern University has developed a structure-mapping engine (SME) that enables computers to reason more like humans and even make moral decisions. The working-process to accomplish this Forbus explains in the following way: “Given a new situation, the machine will try to retrieve one of its prior stories, looking for analogous sacred values, and decide accordingly.” Many artificial intelligence systems, like Google’s AlphaGo, are relying on deep learning of massive amounts of data. While a SME-based system often is learning successfully from far fewer examples e.g. in moral decision-making only a handful of stories. Professor Forbus and his team now is releasing the SME source code and a 5,000-example corpus to encourage further research on analogy, Source: Eurekalert
Latest News: Artificial Intelligence that solves moral dilemmas
Bookmark the permalink.