A simple proposition: There exists no “perfect plan” to guide transitions from legacy (i.e., existing) portfolios to target portfolios. The “good plan” balances art and science, and requires significant resources if it is to be executed quickly and cost effectively. Efficient and cost-effective portfolio transitions require three ingredients: First, sound judgment and extensive experience are n...

Supervisory Team  Enrico Masoero, School of Engineering http://www.ncl.ac.uk/engineering/staff/profile/enricomasoero.ht ml#background  Gerasimos Rigopoulos, School of Mathematics, Statistics, and Physics, http://www.ncl.ac.uk/mathsphysics/staff/profile/gerasimosrigopoulos.html#background  Steve Bull, School of Engineering, http://www.ncl.ac.uk/engineering/staff/profile/stevebull.html#ba ckgr...

In this article we introduce the Arcade Learning Environment (ALE): both a challenge problem and a platform and methodology for evaluating the development of general, domain-independent AI technology. ALE provides an interface to hundreds of Atari 2600 game environments, each one different, interesting, and designed to be a challenge for human players. ALE presents significant research challeng...

The purpose of this research is to analyze the content of e-portfolios created by students in order to understand their tabulation and ways of displaying content. The analytic result shows that the number of outcome portfolios created by students is more than that of process portfolios. The five types of e-portfolio tabulation, in order of those most commonly created by students, are combinatio...

In reinforcement learning (RL) tasks, an efficient exploration mechanism should be able to encourage an agent to take actions that lead to less frequent states which may yield higher accumulative future return. However, both knowing about the future and evaluating the frequentness of states are non-trivial tasks, especially for deep RL domains, where a state is represented by high-dimensional i...

L. M. ROBLEDO, M. WARDA a Departamento de F́ısica Teórica C-XI, Universidad Autónoma de Madrid, 28-049 Madrid, Spain b Departament d’Estructura i Constituents de la Matèria and Institut de Ciències del Cosmos, Facultat de F́ısica, Universitat de Barcelona, Diagonal 647, 08028 Barcelona, Spain c Katedra Fizyki Teoretycznej, Uniwersytet Marii Curie–Sk lodowskiej, ul. Radziszewskiego 10, 20-031 Lubl...

Michael Stich, Josep M. Ribó and David Hochberg Non-linearity and Complexity Research Group, School of Engineering and Applied Science, Aston University, B4 7ET Birmingham, UK Department of Organic Chemistry, Institute of Cosmos Science (IEEC-UB), University of Barcelona, Barcelona, Spain and Department of Molecular Evolution, Centro de Astrobioloǵıa (CSIC-INTA), Carretera Ajalvir Kilómetro 4, ...

We uncover the origin of the puzzling anti-correlation between the formation epoch of galactic dark-matter haloes and their environment density. This correlation has been revealed from cosmological N -body simulations and it is in conflict with the simple excursion-set model of halo clustering. Using similar simulations, we first quantify the straightforward association of an early formation ep...

Recent work in deep reinforcement learning has allowed algorithms to learn complex tasks such as Atari 2600 games just from the reward provided by the game, but these algorithms presently require millions of training steps in order to learn, making them approximately five orders of magnitude slower than humans. One reason for this is that humans build robust shared representations that are appl...