FURI | Spring 2020

Is Artificial Intelligence a Burden in a Changing Environment

Although there have been many recent advancements in artificial intelligence (A.I.) learning, the capacity to make assumptions is necessary to reduce the time, data, and error it takes in order to approach human learning. An environment that changes enough so small patterns must be predicted to optimally learn is required to promote assumptive behavior. Building upon a colleague’s virtual environment, and testing with DeepMind’s OpenSpiel framework, an environment of this caliber has begun development. Subsequent research on assumptive A.I. learning could benefit from this environment to train and measure how much the A.I. helps a human achieve a desired goal.