We propose a Bayesian nonparametric approach to the problem of jointly modeling multiple related time series. Our approach is based on the discovery of a set of latent, shared dynamical behaviors. Using a beta process prior, the size of the set and the sharing pattern are both inferred from data. We develop efficient Markov chain Monte Carlo methods based on the Indian buffet process representation of the predictive distribution of the beta process, without relying on a truncated model. In particular, our approach uses the sum-product algorithm to efficiently compute Metropolis-Hastings acceptance probabilities, and explores new dynamical behaviors via birth and death proposals. We examine the benefits of our proposed feature-based model on several synthetic datasets, and also demonstrate promising results on unsupervised segmentation of visual motion capture data.

This paper illustrates successful implementation of three evolutionary algorithms, namely- Particle Swarm Optimization(PSO), Artificial Bee Colony (ABC) and Bacterial Foraging Optimization (BFO) algorithms to economic load dispatch problem (ELD). Power output of each generating unit and optimum fuel cost obtained using all three algorithms have been compared. The results obtained show that ABC and BFO algorithms converge to optimal fuel cost with reduced computational time when compared to PSO for the two example problems considered.

The purpose of this paper is to investigate the problem of healing of cracks in infinite thermo elastic surroundings by currents of fluids with inclusions. The process of healing is result of growing of layer of sediments on crack’s surfaces due to transverse diffusion currents of crystalline inclusions which leads to closing of crack. The obtained analytical formulae for vertical displacements of boundary of crack and graphs of it allow determine coordinates and time moments of zero width of crack, i.e. process of healing of crack. The applications to biological and technological problems when as mixture is used fluid with camphor crystalline as well as to geothermal cracks with fluid currents containing silicon crystalline with application to geophysics, and results of calculations also can be used in meso mechanics. The investigation of obtained curves of dependence of times and coordinates of healing of cracks processes is done as by deterministic treatment, as well as on account of possible randomness, taking places for micro sizes of cracks by methods of nonlinear wave dynamics. The main methods of investigation of healing problem for cracks are taking into account of thermo diffusion and thermo elasticity stresses and displacements effects and trybological supplements in boundary condition on the cracks., and solution of posed problem of thermo elasticity by method of integral transformations of Laplace and Fourier, method of Winner-Hopf. The analytical solution for displacements of mentioned unsteady plane problem is brought to effective Smirnov-Sobolev form. The calculations of integrals for displacements on crack surface are carried out for number of typical constants of media and parameters of biomechanical, technological, geophysical, nanophysics problems. Furthermore there are carried out applications of methods of nonlinear wave dynamics to finding of probabilities of processes of healing of cracks.These analytic and numerical methods based on dynamic thermo-elasticity approximation on account of diffusion currents of inclusions in fluid mixture in cracks allow determine time and coordinate of closing of cracks. Besides are examined obtained curves of healing processes by methods of nonlinear waves dynamics. These methods can be applied also to extremely processes of transition from meso level defects to synchronized processes of generation of macro cracks and fracture.

This research aims to define a novel framework to employ engineering and mathematical models to study adaptive dynamics in heterarchial systems. This multi-profile descriptive platform and modeling approach is developed as a composite of conceptual behaviors and structural entity aspects of engineered complex adaptive systems (ECAS). While the US electric power system will be utilized for demonstration and validation, the framework has applicability to the general class of ECASs that are artificially created but highly interactive with natural and behavioral sciences. Conditioned on parameterization of the framework, a theorem will be presented to calibrate current structure and predict future dynamic behaviors of an ECAS. We analyze decentralized heterarchial ECASs to infer emergent behavior of the components, and evolution processes and adaptations of the whole system.

In this paper we develop a model to simulate the evolution of a pollution-free resource gathering technology that is initially less efficient but ultimately reaches parity with polluting technology. We find that for low levels of pollution, pressure exerted by society can indeed encourage the development and use of non-polluting technology, with greater pressure being associated with faster achievement of efficiency parity and lower overall pollution. However, greater pressure is also associated with lower populations and at the highest levels of pressure there are significant risks of population crashes. We find that these results hold for both localized pollution and globalized pollution, with globalized pollution encouraging faster achievement of efficiency parity. For high levels of pollution we find that introducing societal pressure significantly increases the occurrence of population crashes, and thus the strategy is only effective under certain conditions.

Broadly speaking, information has something to do with order or organization within a system of elements. The thermodynamic concept of entropy is also associated with such systems, although in an inverse relationship. When we attempt to put these two apparently coordinated schemas of order and disorder together, all kinds of difficulties arise. I will briefly examine contemporary efforts to unify these two ways of conceiving order and show that they are substantially incompatible. In this process I will draw some distinctions that will lead to a broader reconciliation of the concepts of order and information. I will then attempt to re-evaluate the fundamental models behind these dissonant traditions for formulating order in an attempt to reframe a synthesis of conceptual structures that are mutually reconcilable. I will try to show that such a synthesis can finally make sense of the stubborn inconsistencies that persist in the ways Newtonian dynamics, thermodynamics and biology utilize the implicitly conflicting arrows of time.

The new scenarios of contemporary adaptive robotics seem to suggest a transformation of the traditional methods. In the search for new approaches to the control of adaptive autonomous systems, the mind becomes a fundamental source of inspiration. In this paper we anticipate, through the use of simulation, the cognitive and behavioral properties that emerge from a recent prototype robotic platform, EcoBot, a family of bio-mechatronic symbionts provided with an `artificial metabolism', that has been under physical development during recent years. Its energy reliance on a biological component and the consequent limitation of its supplied energy determine a special kind of dynamic coupling between the robot and its environment. Rather than just an obstacle, energetic constraints become the opportunity for the development of a rich set of behavioral and cognitive properties.

Inference techniques play a central role in many cognitive systems. They transform low-level observations of the environment into high-level, actionable knowledge which then gets used by mechanisms that drive action, problem-solving, and learning. This paper presents an initial effort at combining results from AI and psychology into a pragmatic and scalable computational reasoning system. Our approach combines a numeric notion of plausibility with first-order logic to produce an incremental inference engine that is guided by heuristics derived from the psychological literature. We illustrate core ideas with detailed examples and discuss the advantages of the approach with respect to cognitive systems.

This paper proposes a new randomized strategy for adaptive MCMC using Bayesian optimization. This approach applies to non-differentiable objective functions and trades off exploration and exploitation to reduce the number of potentially costly objective function evaluations. We demonstrate the strategy in the complex setting of sampling from constrained, discrete and densely connected probabilistic graphical models where, for each variation of the problem, one needs to adjust the parameters of the proposal mechanism automatically to ensure efficient mixing of the Markov chains.

One of the important issues in computer networks is "Load Balancing" which leads to efficient use of the network resources. To achieve a balanced network it is necessary to find different routes between the source and destination. In the current paper we propose a new approach to find different routes using swarm intelligence techniques and multi colony algorithms. In the proposed algorithm that is an improved version of MACO algorithm, we use different colonies of ants and bees and appoint these colony members as intelligent agents to monitor the network and update the routing information. The survey includes comparison and critiques of MACO. The simulation results show a tangible improvement in the aforementioned approach.