Learning in Network Games

We report the findings of an experiment designed to study how people learn and make decisions in network games. Network games offer new opportunities to identify learning rules, since on networks (compared to e.g. random matching) more rules differ in terms of their information requirements. Our experimental design enables us to observe both which actions participants choose and which information they consult before making their choices. We use this information to estimate learning types using maximum likelihood methods. There is substantial heterogeneity in learning types. However, the vast majority of our participants' decisions are best characterized by reinforcement learning or (myopic) best-response learning. The distribution of learning types seems fairly stable across contexts. Neither network topology nor the position of a player in the network seem to substantially affect the estimated distribution of learning types.

New methods for generating populations in Markov network based EDAs: Decimation strategies and model-based template recombination

Methods for generating a new population are a fundamental component of estimation of distribution algorithms (EDAs). They serve to transfer the information contained in the probabilistic model to the new generated population. In EDAs based on Markov networks, methods for generating new populations usually discard information contained in the model to gain in efficiency. Other methods like Gibbs sampling use information about all interactions in the model but are computationally very costly. In this paper we propose new methods for generating new solutions in EDAs based on Markov networks. We introduce approaches based on inference methods for computing the most probable configurations and model-based template recombination. We show that the application of different variants of inference methods can increase the EDAs’ convergence rate and reduce the number of function evaluations needed to find the optimum of binary and non-binary discrete functions.

MBARI procedures for automated nutrient analyses using a modified Alpkem Series 300 Rapid Flow Analyzer

This report documents the methods used at the Monterey Bay Aquarium Research Institute (MBARI) for analyzing seawater nutrient samples with an Alpkem Series 300 Rapid Flow Analyzer (RFA) system. The methods have been optimized for the particular requirements of this laboratory. The RFA system has been used to analyze approximately 20,000 samples during the past two years. The methods have been optimized to run nutrient analyses in a routine manner with a detection limit of better than -±1% and a within run precision of -±1% of the full scale concentration range. The normal concentration ranges are 0-200 ~M silicate, 0-5 ~M phosphate, 0-50 ~M nitrate, 0-3 ~M nitrite, and 0-10 ~M ammonium. The memorandum is designed to be used in a loose-leaf binder format. Each page is dated and as revisions are made, they should be inserted into the binder. The revisions should be added into the binder. Retain the old versions in order to maintain a historical record of the procedures. (88 pages)

A Comparison of Two Methods for Enhancing the Recovery of Seagrasses into Propellor Scars: Mechanical Injection of a Nutrient and Growth Hormone Solution vs. Defecation by Roosting Seabirds: Final Report.

Based on the recovery rates for Thalassia testudinum measured in this study for scars of these excavation depths and assuming a linear recovery horizon, we estimate that it would take ~ 6.9 years (95% CI. = 5.4 to 9.6 years) for T. testudinum to return to the same density as recorded for the adjacent undisturbed population. The application of water soluble fertilizers and plant growth hormones by mechanical injection into the sediments adjacent to ten propellor scars at Lignumvitae State Botanical Site did not significantly increase the recovery rate of Thalassia testudinum or Halodule wrightii. An alternative method of fertilization and restoration of propellor scars was also tested by a using a method of “compressed succession” where Halodule wrightii is substituted for T. testudinum in the initial stages of restoration. Bird roosting stakes were placed among H.wrightii bare root plantings in prop scars to facilitate the defecation of nitrogen and phosphorus enriched feces. In contrast to the fertilizer injection method, the bird stakes produced extremely high recovery rates of transplanted H. wrightii. We conclude that use of a fertilizer/hormone injection machine in the manner described here is not a feasible means of enhancing T. testudinum recovery in propellor scars on soft bottom carbonate sediments. Existing techniques such as the bird stake approach provide a reliable, and inexpensive alternative method that should be considered for application to restoration of seagrasses in these environments. Document contains 40 pages)

Birkenhead Sixth Form College: Opening up remote access to the college network brings wider benefits to all

Birkenhead Sixth Form College implemented a virtual network to open up remote access to the college network for its students, staff and governors. In particular, for childcare students on work placements, this has meant 24/7 secure access to their work and resources, and the ability to make timely updates to their work evidence logs. The impact is better continuity of learning and a dramatic increase in the hand-in rate for work. For the staff, governors and college as a whole, the benefits of anytime-access to the network are more than were envisaged at the outset; not only is it saving them valuable time and eliminating the need for large print runs, it is expected to bring cost-savings to the College in the long term.

Comparisons of static, quasi-static and dynamic 3D porous media scale network models for two-phase immiscible flow in porous media

A dynamic 3D pore-scale network model is formulated for investigating the effect of interfacial tension and oil-water viscosity during chemical flooding. The model takes into account both viscous and capillary forces in analyzing the impact of chemical properties on flow behavior or displacement configuration, while the static model with conventional invasion percolation algorithm incorporates the capillary pressure only. From comparisons of simulation results from these models. it indicates that the static pore scale network model can be used successfully when the capillary number is low. With the capillary increases due to the enhancement of water viscosity or decrease of interfacial tension, only the quasi-static and dynamic model can give insight into the displacement mechanisms.

IPv6 technical summary for network managers

Aimed at network managers who are responsible for managing the network infrastructure for their organisation and want come understanding of the issues involved in enabling IPv6. JL

Influence of Training Set Selection in Artificial Neural Network-Based Propagation Path Loss Predictions

This paper analyzes the use of artificial neural networks (ANNs) for predicting the received power/path loss in both outdoor and indoor links. The approach followed has been a combined use of ANNs and ray-tracing, the latter allowing the identification and parameterization of the so-called dominant path. A complete description of the process for creating and training an ANN-based model is presented with special emphasis on the training process. More specifically, we will be discussing various techniques to arrive at valid predictions focusing on an optimum selection of the training set. A quantitative analysis based on results from two narrowband measurement campaigns, one outdoors and the other indoors, is also presented.