Learn More
— How can an agent bootstrap up from a pixel-level representation to autonomously learn high-level states and actions using only domain-general knowledge? In this paper we assume that the learning agent has a set of continuous variables describing the environment. There exist methods for learning models of the environment, and there also exist methods for(More)
By allowing individuals to be permanently connected to the Internet, mobile devices ease the way information can be ac-cessed and shared online, but also raise novel privacy challenges for end users. Recent behavioral research on " soft " or " asymmetric " paternalism has begun exploring ways of helping people make better decisions in different aspects of(More)
With smart-phones becoming increasingly commonplace, there has been a subsequent surge in applications that continuously track the location of users. However, serious privacy concerns arise as people start to widely adopt these applications. Users will need to maintain policies to determine under which circumstances to share their location. Specifying these(More)
There has been intense interest in hierarchical reinforcement learning as a way to make Markov decision process planning more tractable, but there has been relatively little work on autonomously learning the hierarchy, especially in continuous domains. In this paper we present a method for learning a hierarchy of actions in a continuous environment. Our(More)
We present a method that allows an agent through active exploration to autonomously build a useful representation of its environment. The agent builds the representation by iteratively learning distinctions and predictive rules using those distinctions. We build on earlier work in which we showed that by motor babbling an agent could learn a representation(More)
As mobile and social networking applications continue to proliferate, they also increasingly rely on the collection of an ever wider range of con-textual attributes, location being a prime example. Prior work has shown that people's privacy preferences when it comes to sharing this information are often complex and that expecting users to spend the time(More)
Frequently one wants to extend the use of a classification method that in principle requires records with True/False values, so that records with rational numbers can be processed. In such cases, the rational numbers must first be replaced by True/False values before the method may be applied. In other cases, a classification method in principle can process(More)
We consider the problem of how a learning agent in a continuous and dynamic world can autonomously learn about itself, its environment , and how to perform simple actions. In previous work we showed how an agent could learn an abstraction consisting of contingencies and distinctions. In this paper we propose a method whereby an agent using this abstraction(More)
— A developing agent must learn the structure of its world, beginning with its sensorimotor world. It learns rules to predict how its motor signals change the sensory input it receives. It learns the limits to its motion. It learns which effects of its actions are unconditional and which effects are conditional, including what they depend on. We present(More)
To my family Acknowledgments I would first like to thank my advisor Ben Kuipers for spending uncountable hours over the last six years to teach me how to be a scientist and how to be precise in my speech and writing. I would also like to thank my committee members. Dana Ballard, for teaching me that if a problem is hard, then maybe there is a simple(More)