Search decisions are often made using heuristic methods because realworld applications can rarely be tackled without any heuristics. In many cases, multiple heuristics can potentially be chosen, and it is not clear a priori which would perform best. In this article, we propose a procedure that learns, during the search process, how to select promising heuristics. The learning is based on weight...

In many applications, observed spatial variables have skewed distributions. It is often of interest to model the shape of the skewed marginal distribution as well as the spatial correlations. We propose a class of stationary processes that have skewed marginal distributions. The covariance function of the process can be given explicitly. We study maximum likelihood inference through a Monte Car...

Learning in the mobile robot domain is a very challenging task, especially in non-stationary conditions. This paper presents an approach that allows a robot to learn a model of its interactions with its operating environment in order to manage them according to the experienced dynamics. The robot is initially given a set of \behavior-producing" modules to choose from, and the algorithm provides...

For a supercritical branching process (Zn) in a stationary and ergodic environment ξ, we study the rate of convergence of the normalized population Wn = Zn/E[Zn|ξ] to its limitW∞: we show a central limit theorem forW∞−Wn with suitable normalization and derive a Berry-Esseen bound for the rate of convergence in the central limit theorem when the environment is independent and identically distrib...

Traditionally, studies of the visual system of nonhuman primates have investigated neurons while the animal fixates a target in a static environment. Clearly, this is not what our everyday life is like; neither the environment nor we are stationary while we act in our world. On the contrary, we constantly move our eyes or limbs. Here we review a number of recent studies describing the brain in ...

Reinforcement learning induces non-stationarity at several levels. Adaptation to non-stationary environments is of course a desired feature of a fair RL algorithm. Yet, even if the environment of the learning agent can be considered as stationary, generalized policy iteration frameworks, because of the interleaving of learning and control, will produce non-stationarity of the evaluated policy a...

We show that in the canonical non-cooperative multilateral bargaining game, a subgame perfect equilibrium exists in pure stationary strategies, even when the space of feasible payoffs is not convex. At such an equilibrium there is no delay. We also have the converse result that randomization will not be used in this environment in the sense that all stationary subgame perfect equilibria do not ...

Speed profiles for motor vehicles on urban streets are influenced not only by the street design and traffic performance, events such as interactions with pedestrians and cyclists, buses, on-street parking, exits from minor roads etc. (called “side-friction events”) also affect driver behaviour. A bottom-up research methodology was developed and applied including a wide range of stationary and m...