We introduce and analyze a new algorithm for linear classification which combines Rosenblatt's perceptron algorithm with Helmbold and Warmuth's leave-one-out method. Like...
Dynamic Programming, Q-learning and other discrete Markov Decision Process solvers can be applied to continuous d-dimensional state-spaces by quantizing the state space into an arr...
Software components for distance and just-in-time (JIT) learning are an increasingly common method of encouraging reuse and facilitating the development process[58], but no analog...
Quite a bit is known about minimizing different kinds of regret in experts problems, and how these regret types relate to types of equilibria in the multiagent setting of repeated...
We propose a fast batch learning method for linearchain Conditional Random Fields (CRFs) based on Newton-CG methods. Newton-CG methods are a variant of Newton method for high-dime...
Yuta Tsuboi, Yuya Unno, Hisashi Kashima, Naoaki Ok...