This paper shows how a text classifier's need for labeled training documents can be reduced by taking advantage of a large pool of unlabeled documents. We modify the Query-by...
This paper introduces the RL-TOPs architecture for robot learning, a hybrid system combining teleo-reactive planning and reinforcement learning techniques. The aim of this system ...
This paper discusses the unsupervised learning problem. An important part of the unsupervised learning problem is determining the numberofconstituent groups (componentsor classes)...
Jonathan J. Oliver, Rohan A. Baxter, Chris S. Wall...
One of the major global problems is bringing learning to everyone on earth, at all ages. Current lockstep cookie-cuter approaches will not accomplish this, because students differ...
In this paper we derive convergence rates for Q-learning. We show an interesting relationship between the convergence rate and the learning rate used in Q-learning. For a polynomi...