This paper considers whether lottery betting is best explained by risk-love or an alternative to the expected utility model, namely, the overweighting of small odds. To the best of our knowledge, we are the first to use the common setting of state lottery betting to investigate the fit of expected utility theory against an alternative model of non-linear probability weighting. Our results show ...

This article addresses weighting and partitioning, in complex reinforcement learning tasks, with the aim of facilitating learning. The article presents some ideas regarding weighting of multiple agents and extends them into partitioning an input/state space into multiple regions with differential weighting in these regions, to exploit differential characteristics of regions and differential cha...

The context-tree weighting method (Willems, Shtarkov, and Tjalkens [1995]) is a sequential universal source coding method that achieves the Rissanen lower bound [1984] for tree sources. The same authors also proposed context-tree maximizing, a two-pass version of the context-tree weighting method [1993]. Later Willems and Tjalkens [1998] described a method based on ratios (betas) of sequence pr...

Term Weighting (TW) is one of the most important tasks for Information Retrieval (IR). To solve the TW problem, many authors have considered Vector Space Model, and specifically, they have used the TF-IDF method. As this method does not take into account some of the features of terms, we propose a novel alternative fuzzy logic based method for TW in IR. TW is an essential task for the Web Intel...

Contrast-enhanced MRA methods which provide temporal information, such as the 3D-TRICKS technique[1], have tradeoffs between temporal resolution and spatial resolution[2]. Acquiring data with adequate temporal resolution often limits the achievable spatial resolution. It is desirable to have high in-plane resolution along with good temporal resolution. Recently we have shown that angularly unde...

Reinforcement Learning (RL) can model complex behavior policies for goaldirected sequential decision making tasks. A hallmark of RL algorithms is Temporal Difference (TD) learning: value function for the current state is moved towards a bootstrapped target that is estimated using next state’s value function. λ-returns generalize beyond 1-step returns and strike a balance between Monte Carlo and...

This letter describes a means to produce emergent collaboration between human driver and an artificial co-driver agent. The work exploits the hypothesis that human-human cooperation emerges from shared understanding of given context’s affordances emulates same principle: observation one agent’s behavior steers another decision-making by favoring selection goals would observed activity. Specific...

This paper looks at unobserved components models and examines the implied weighting patterns for signal extraction. There are three main themes. The rst is the implications of correlated disturbances driving the components, especially those cases in which the correlation is perfect. The second is how setting up models with t distributed disturbances leads to weighting patterns which are robust ...

First we modify the basic (binary) context-tree weighting method such that the past symbols x1 D; x2 D; ; x0 are not needed by the encoder and the decoder. Then we describe how to make the context-tree depth D infinite, which results in optimal redundancy behavior for all tree sources, while the number of records in the context tree is not larger than 2T 1: Here T is the length of the source se...