Stuart I. Reynolds sir s.bham.a .uk S hool of Computer S ien e, The University of Birmingham, Birmingham, B15 2TT, UK 1. The max Operator and Optimisti Initial Biases This paper introdu es the following simple insight: RL algorithms that update their value estimates based upon a return estimate involving maxaQ(s; a) nd it more diÆ ult to overome their initial value biases if these biases are op...

In this text, some question related to higher dimensional geometrical spaces will be discussed. The goal is to give the reader a feeling for geometric distortions related to the use of such spaces (e.g. as search spaces).

Despite its enormous promise to further our understanding of cellular processes involved in the regulation of gene expression, microarray technology generates data for which statistical pre-processing has become a necessity before any interpretation of data can begin. The process by which we distinguish (and remove) non-biological variation from biological variation is called normalization. Wit...

This paper proposes a model that explains the nonneutrality of money from two welldocumented psychological assumptions. The model incorporates into the general-equilibrium monopolistic-competition framework of Blanchard and Kiyotaki [1987] the psychological assumptions that (1) consumers dislike paying a price that exceeds some “fair” markup on firms’ marginal costs, and (2) consumers do not kn...

The previous lecture defined the Nearest Neighbor Algorithm and discussed how it suffers from the curse of dimensionality. This means that as the number of dimensions increase the Nearest Neighbor algorithm performs poorer and poorer. To better understand the curse of dimensionality with regard to the Nearest Neighbor algorithm one must understand what higher dimensions look like. The following...