Sometraditional models [4,24,25,26,27,29,30] rely on the assumptions and heuristics rules about various measurements of image similarity to perform abstract reasoning.

We give a model of how to infer natural language rules by doing experiments.

Zero-sum games can be solved using online learning dynamics, where a classical technique involves simulating two no-regret algorithms that play against each other and, afterT rounds, the average iterate is guaranteed to solve the original optimization problem with error decaying asO(logT/T). In this paper we show that the technique can be enhanced to a rate ofO(1/T2) by extending recent work [22, 25] that leverages optimistic learning to speed upequilibrium computation.

Neural Information Processing Systems http://nips.cc/

Haussler et al. [1994], Ryabko [2006] researched the online learning problem with a mix of both restrictions. There are also substantial amount of papers investigating online learnability with all measurable functions but restricted data processes.

In this paper, we propose an online convex optimization approach with two different levels of adaptivity. On a higher level, our approach is agnostic to the unknown types and curvatures of the online functions, while at a lower level, it can exploit the unknown niceness of the environments and attain problem-dependent guarantees.