Optimal management of beaver population using a reduced-order distributed parameter model and single network adaptive critics

Beavers are often found to be in conflict with human interests by creating nuisances like building dams on flowing water (leading to flooding), blocking irrigation canals, cutting down timbers, etc. At the same time they contribute to raising water tables, increased vegetation, etc. Consequently, maintaining an optimal beaver population is beneficial. Because of their diffusion externality (due to migratory nature), strategies based on lumped parameter models are often ineffective. Using a distributed parameter model for beaver population that accounts for their spatial and temporal behavior, an optimal control (trapping) strategy is presented in this paper that leads to a desired distribution of the animal density in a region in the long run. The optimal control solution presented, imbeds the solution for a large number of initial conditions (i.e., it has a feedback form), which is otherwise nontrivial to obtain. The solution obtained can be used in real-time by a nonexpert in control theory since it involves only using the neural networks trained offline. Proper orthogonal decomposition-based basis function design followed by their use in a Galerkin projection has been incorporated in the solution process as a model reduction technique. Optimal solutions are obtained through a "single network adaptive critic" (SNAC) neural-network architecture.

Locating and sizing plants for bottling propane in south India

Shri Shakti LPG Ltd. (SSLPG) imports and markets propane (referred to as liquefied petroleum gas (LPG) in India) in south India. It sells LPG in packed (cylinder) form to domestic customers and commercial establishments through a network of dealers. Dealers replenish their stocks of filled cylinders from bottling plants, which in turn receive LPG in bulk from the cheaper of SSLPG's two import-and-storage facilities that are located on the Indian coast. We implemented integer programming to help SSLPG decide on the locations and long-run sizes of its bottling plants. We estimate that our recommended configuration of bottling plants is about $1 million cheaper annually than the one that SSLPG had initially planned.

Temperature dependence of the levitation effect - Implications for separation of multicomponent mixtures

Sufficiently long molecular dynamics simulations have been carried out on spherical monatomic sorbates in NaY zeolite, interacting via simple Lennard-Jones potentials, to investigate the dependence of the levitation effect on the temperature. Simulations carried out in the range 100-300 K suggest that the anomalous peak in the diffusion coefficient (observed when the levitation parameter, gamma, is near unity) decreases in intensity with increase in temperature. The rate of cage-to-cage migrations also exhibits a similar trend. The activation energy obtained from Arrhenius plots is found to exhibit a minimum when the diffusion coefficient is a maximum, corresponding to the gamma approximate to 1 sorbate diameter. In the linear or normal regime, the activation energy increases with increase in sorbate diameter until it shows a sharp decrease in the anomalous regime. Locations and energies of the adsorption sites and their dependence on the sorbate size gives interesting insight into the nature of the underlying potential-energy surface and further explain the observed trend in the activation energy with sorbate size. Cage residence times, tau(c), show little or no change with temperature for the sorbate with diameter corresponding to gamma approximate to 1, whereas there is a significant decrease in tau(c) with increase in temperature for sorbates in the linear regime. The implications of the present study for the separation of mixtures of sorbates are discussed.

Stochastic Algorithms for Discrete Parameter Simulation Optimization

We present two efficient discrete parameter simulation optimization (DPSO) algorithms for the long-run average cost objective. One of these algorithms uses the smoothed functional approximation (SFA) procedure, while the other is based on simultaneous perturbation stochastic approximation (SPSA). The use of SFA for DPSO had not been proposed previously in the literature. Further, both algorithms adopt an interesting technique of random projections that we present here for the first time. We give a proof of convergence of our algorithms. Next, we present detailed numerical experiments on a problem of admission control with dependent service times. We consider two different settings involving parameter sets that have moderate and large sizes, respectively. On the first setting, we also show performance comparisons with the well-studied optimal computing budget allocation (OCBA) algorithm and also the equal allocation algorithm. Note to Practitioners-Even though SPSA and SFA have been devised in the literature for continuous optimization problems, our results indicate that they can be powerful techniques even when they are adapted to discrete optimization settings. OCBA is widely recognized as one of the most powerful methods for discrete optimization when the parameter sets are of small or moderate size. On a setting involving a parameter set of size 100, we observe that when the computing budget is small, both SPSA and OCBA show similar performance and are better in comparison to SFA, however, as the computing budget is increased, SPSA and SFA show better performance than OCBA. Both our algorithms also show good performance when the parameter set has a size of 10(8). SFA is seen to show the best overall performance. Unlike most other DPSO algorithms in the literature, an advantage with our algorithms is that they are easily implementable regardless of the size of the parameter sets and show good performance in both scenarios.

An Online Actor-Critic Algorithm with Function Approximation for Constrained Markov Decision Processes

We develop an online actor-critic reinforcement learning algorithm with function approximation for a problem of control under inequality constraints. We consider the long-run average cost Markov decision process (MDP) framework in which both the objective and the constraint functions are suitable policy-dependent long-run averages of certain sample path functions. The Lagrange multiplier method is used to handle the inequality constraints. We prove the asymptotic almost sure convergence of our algorithm to a locally optimal solution. We also provide the results of numerical experiments on a problem of routing in a multi-stage queueing network with constraints on long-run average queue lengths. We observe that our algorithm exhibits good performance on this setting and converges to a feasible point.

Virtually cool ternary content addressable memory

Fast content addressable data access mechanisms have compelling applications in today's systems. Many of these exploit the powerful wildcard matching capabilities provided by ternary content addressable memories. For example, TCAM based implementations of important algorithms in data mining been developed in recent years; these achieve an an order of magnitude speedup over prevalent techniques. However, large hardware TCAMs are still prohibitively expensive in terms of power consumption and cost per bit. This has been a barrier to extending their exploitation beyond niche and special purpose systems. We propose an approach to overcome this barrier by extending the traditional virtual memory hierarchy to scale up the user visible capacity of TCAMs while mitigating the power consumption overhead. By exploiting the notion of content locality (as opposed to spatial locality), we devise a novel combination of software and hardware techniques to provide an abstraction of a large virtual ternary content addressable space. In the long run, such abstractions enable applications to disassociate considerations of spatial locality and contiguity from the way data is referenced. If successful, ideas for making content addressability a first class abstraction in computing systems can open up a radical shift in the way applications are optimized for memory locality, just as storage class memories are soon expected to shift away from the way in which applications are typically optimized for disk access locality.

Transmit power Control with ARQ in energy harvesting sensors: a decision-theoretic approach

This paper addresses the problem of finding optimal power control policies for wireless energy harvesting sensor (EHS) nodes with automatic repeat request (ARQ)-based packet transmissions. The EHS harvests energy from the environment according to a Bernoulli process; and it is required to operate within the constraint of energy neutrality. The EHS obtains partial channel state information (CSI) at the transmitter through the link-layer ARQ protocol, via the ACK/NACK feedback messages, and uses it to adapt the transmission power for the packet (re)transmission attempts. The underlying wireless fading channel is modeled as a finite state Markov chain with known transition probabilities. Thus, the goal of the power management policy is to determine the best power setting for the current packet transmission attempt, so as to maximize a long-run expected reward such as the expected outage probability. The problem is addressed in a decision-theoretic framework by casting it as a partially observable Markov decision process (POMDP). Due to the large size of the state-space, the exact solution to the POMDP is computationally expensive. Hence, two popular approximate solutions are considered, which yield good power management policies for the transmission attempts. Monte Carlo simulation results illustrate the efficacy of the approach and show that the approximate solutions significantly outperform conventional approaches.

Adaptive Sleep-Wake Control using Reinforcement Learning in Sensor Networks

The aim in this paper is to allocate the `sleep time' of the individual sensors in an intrusion detection application so that the energy consumption from the sensors is reduced, while keeping the tracking error to a minimum. We propose two novel reinforcement learning (RL) based algorithms that attempt to minimize a certain long-run average cost objective. Both our algorithms incorporate feature-based representations to handle the curse of dimensionality associated with the underlying partially-observable Markov decision process (POMDP). Further, the feature selection scheme used in our algorithms intelligently manages the energy cost and tracking cost factors, which in turn assists the search for the optimal sleeping policy. We also extend these algorithms to a setting where the intruder's mobility model is not known by incorporating a stochastic iterative scheme for estimating the mobility model. The simulation results on a synthetic 2-d network setting are encouraging.

Energy Sharing for Multiple Sensor Nodes With Finite Buffers

We consider the problem of finding optimal energy sharing policies that maximize the network performance of a system comprising of multiple sensor nodes and a single energy harvesting (EH) source. Sensor nodes periodically sense the random field and generate data, which is stored in the corresponding data queues. The EH source harnesses energy from ambient energy sources and the generated energy is stored in an energy buffer. Sensor nodes receive energy for data transmission from the EH source. The EH source has to efficiently share the stored energy among the nodes to minimize the long-run average delay in data transmission. We formulate the problem of energy sharing between the nodes in the framework of average cost infinite-horizon Markov decision processes (MDPs). We develop efficient energy sharing algorithms, namely Q-learning algorithm with exploration mechanisms based on the epsilon-greedy method as well as upper confidence bound (UCB). We extend these algorithms by incorporating state and action space aggregation to tackle state-action space explosion in the MDP. We also develop a cross entropy based method that incorporates policy parameterization to find near optimal energy sharing policies. Through simulations, we show that our algorithms yield energy sharing policies that outperform the heuristic greedy method.

Capacidad de crecimiento de la Argentina en el largo plazo

Resumen: En los últimos cuatro años la Argentina creció 39% aproximadamente. Varios factores se conjugaron para permitir este crecimiento, pero sin duda, uno clave fue la existencia de una importante capacidad productiva ociosa, algo que se está diluyendo. En este contexto es importante conocer cuánto margen hay para crecer en el corto plazo, o sea, cuánto es el output gap o brecha del producto, y a qué ritmo se puede crecer en el largo plazo, o sea, la tasa de crecimiento del PBI potencial. Para ello en el presente artículo se calcula el PBI potencial en base al “método de la función de producción”. La principal conclusión es que para crecer a 4% en el largo plazo, una tasa mucho menor a la registrada recientemente, se requiere un esfuerzo inversor mayor al actual, tanto en cantidad como en calidad.

Las reformas económicas y la modernización del Estado

Resumen: El presente artículo trata conjuntamente acerca de los contenidos y la estrategia para la Modernización del Estado que se desprende de las recientes experiencias en cuanto a la aplicación de las Reformas Económicas de los años 80s y 90s, como así también de los últimos desarrollos de nuevas áreas de investigación como la Economía Institucional y la Teoría del Capital Social. Desde este punto de vista para diseñar una Estrategia de Modernización del Estado es esencial una visión estratégica de las actividades que el Estado debe realizar, delegar o incentivar. Asimismo las orientaciones fundamentales pasan por la existencia de consensos sobre políticas de largo plazo, la independencia del funcionamiento, un correcto diseño de reglas en lugar de conductas discrecionales, un balance entre subsidiariedad y participación, el logro de la excelencia de la función pública y de la lucha contra la corrupción.

Turnout Intention and Social Networks

How can networking affect the turnout in an election? We present a simple model to explain turnout as a result of a dynamic process of formation of the intention to vote within Erdös-Renyi random networks. Citizens have fixed preferences for one of two parties and are embedded in a given social network. They decide whether or not to vote on the basis of the attitude of their immediate contacts. They may simply follow the behavior of the majority (followers) or make an adaptive local calculus of voting (Downsian behavior). So they either have the intention of voting when the majority of their neighbors are willing to vote too, or they vote when they perceive in their social neighborhood that elections are "close". We study the long run average turnout, interpreted as the actual turnout observed in an election. Depending on the combination of values of the two key parameters, the average connectivity and the probability of behaving as a follower or in a Downsian fashion, the system exhibits monostability (zero turnout), bistability (zero turnout and either moderate or high turnout) or tristability (zero, moderate and high turnout). This means, in particular, that for a wide range of values of both parameters, we obtain realistic turnout rates, i.e. between 50% and 90%.

Monte Carlo Valuation of natural gas investments

This paper deals with the valuation of energy assets related to natural gas. In particular, we evaluate a baseload Natural Gas Combined Cycle (NGCC) power plant and an ancillary instalation, namely a Liquefied Natural Gas (LNG) facility, in a realistic setting; specifically, these investments enjoy a long useful life but require some non-negligible time to build. Then we focus on the valuation of several investment options again in a realistic setting. These include the option to invest in the power plant when there is uncertainty concerning the initial outlay, or the option's time to maturity, or the cost of CO2 emission permits, or when there is a chance to double the plant size in the future. Our model comprises three sources of risk. We consider uncertain gas prices with regard to both the current level and the long-run equilibrium level; the current electricity price is also uncertain. They all are assumed to show mean reversion. The two-factor model for natural gas price is calibrated using data from NYMEX NG futures contracts. Also, we calibrate the one-factor model for electricity price using data from the Spanish wholesale electricity market, respectively. Then we use the estimated parameter values alongside actual physical parameters from a case study to value natural gas plants. Finally, the calibrated parameters are also used in a Monte Carlo simulation framework to evaluate several American-type options to invest in these energy assets. We accomplish this by following the least squares MC approach.

Do experimental subjects favor their friends?

Ideally we would like subjects of experiments to be perfect strangers so that the situation they face at the lab is not just part of a long run interaction. Unfortunately, it is not easy to reach those conditions and experimenters try to mitigate any effects from those out-of-the-lab relationships by, for instance, randomly matching subjects. However, even if this type of procedure is used, there is a positive probability that a subject may face a friend or an acquaintance. We find evidence that social proximity between subjects is irrelevant to experiment results in dictator games. Thus, although ideal conditions are not met, relations between subjects do not contaminate the results of experiments.

El papel de internet en la convergencia cultural: un análisis teórico basado en la aproximación constructivista dinámica desde la perspectiva de las interacciones online de los consumidores

[ES] Este trabajo plantea y analiza, con un enfoque teórico, el papel que puede tener Internet como factor impulsor clave de la convergencia cultural entre países. En este respecto, se reflexiona en torno a una propuesta teórica central, tomando como base principal la aproximación cultural constructivista dinámica. En esencia, se teoriza sobre cómo los valores compartidos por parte de los individuos (consumidores) procedentes de diferentes culturas en este medio, fruto de procesos interactivos e iterativos de comunicación online, pueden ser transferidos de manera plausible a cada una de sus culturas primarias o de origen, fomentando, por tanto, la aproximación de las mismas en el largo plazo.