Humans are remarkable in their ability to rapidly learn complex tasks from little experience. Recent successes in AI have produced algorithms that can perform complex tasks well in environments whose simple dynamics are known in advance (e.g., AlphaGo, Libratus), as well as models that can learn to perform expertly in unknown environments after a great amount of experience (e.g., DQN, DDQN, A3C, and so on). Despite this, no current AI models learn sufficiently rich and general representations so as to support rapid, human-level learning on new, complex, tasks.
This thesis examines some of the representations, algorithms, and epistemic practices that may underlie humans’ ability to quickly learn about their world and deploy that understanding to achieve their aims. In particular, the thesis examines humans’ ability to effectively query their environment for information that helps distinguish between competing hypotheses (Chapter 2); children’s ability to use higher-level amodal features of data to match causes and effects (Chapter 3); and adult human rapid-learning abilities in artificial video-game environments (Chapter 4). The thesis culminates by presenting and testing a model, inspired by human inductive biases and epistemic practices, that learns to perform complex video-game tasks at human levels with human-level amounts of experience (Chapters 5 and 6). Our model demonstrates the benefits of incorporating richly structured theories into more traditional model-based RL approaches, and suggests that such “theory-based RL” may offer a promising approach to developing artificial agents that learn as rapidly and flexibly as humans.
View this thesis here: https://www.dropbox.com/s/ory1zvc3khczy3t/Theory_Based_Learning_in_Humans_and_Machines.pdf?dl=0