The road to chaos by time-asymmetric Hebbian learning in recurrent neural networks

This letter aims at studying the impact of iterative Hebbian learning algorithms on the recurrent neural network's underlying dynamics. First, an iterative supervised learning algorithm is discussed. An essential improvement of this algorithm consists of indexing the attractor information items by means of external stimuli rather than by using only initial conditions, as Hopfield originally proposed. Modifying the stimuli mainly results in a change of the entire internal dynamics, leading to an enlargement of the set of attractors and potential memory bags. The impact of the learning on the network's dynamics is the following: the more information to be stored as limit cycle attractors of the neural network, the more chaos prevails as the background dynamical regime of the network. In fact, the background chaos spreads widely and adopts a very unstructured shape similar to white noise. Next, we introduce a new form of supervised learning that is more plausible from a biological point of view: the network has to learn to react to an external stimulus by cycling through a sequence that is no longer specified a priori. Based on its spontaneous dynamics, the network decides "on its own" the dynamical patterns to be associated with the stimuli. Compared with classical supervised learning, huge enhancements in storing capacity and computational cost have been observed. Moreover, this new form of supervised learning, by being more "respectful" of the network intrinsic dynamics, maintains much more structure in the obtained chaos. It is still possible to observe the traces of the learned attractors in the chaotic regime. This complex but still very informative regime is referred to as "frustrated chaos."     

A layered approach to learning coordination knowledge in multiagent environments

Multiagent learning involves acquisition of cooperative behavior among intelligent agents in order to satisfy the joint goals. Reinforcement Learning (RL) is a promising unsupervised machine learning technique inspired from the earlier studies in animal learning. In this paper, we propose a new RL technique called the Two Level Reinforcement Learning with Communication (2LRL) method to provide cooperative action selection in a multiagent environment. In 2LRL, learning takes place in two hierarchical levels; in the first level agents learn to select their target and then they select the action directed to their target in the second level. The agents communicate their perception to their neighbors and use the communication information in their decision-making. We applied 2LRL method in a hunter-prey environment and observed a satisfactory cooperative behavior. Guray Erus received the B.S. degree in computer engineering in 1999, and the M.S. degree in cognitive sciences, in 2002, from Middle East Technical University (METU), Ankara, Turkey. He is currently a teaching and research assistant in Rene“ Descartes University, Paris, France, where he prepares a doctoral dissertation on object detection on satellite images, as a member of the intelligent perception systems group (SIP-CRIP5). His research interests include multi-agent systems and image understanding. Faruk Polat is a professor in the Department of Computer Engineering of Middle East Technical University, Ankara, Turkey. He received his B.Sc. in computer engineering from the Middle East Technical University, Ankara, in 1987 and his M.S. and Ph.D. degrees in computer engineering from Bilkent University, Ankara, in 1989 and 1993, respectively. He conducted research as a visiting NATO science scholar at Computer Science Department of University of Minnesota, Minneapolis in 1992–93. His research interests include artificial intelligence, multi-agent systems and object oriented data models.     

Effect of carbohydrate supplementation provided through drinking water during feed withdrawal on meat and liver colours in broilers

BACKGROUND: The effect of saccarose supplementation provided through drinking water during feed withdrawal on meat and liver colour was investigated in a total of 50 broilers (five floor pens with 10 broilers) at 43 days. Treatments consisted of: (1) full‐fed control broilers fed the standard broiler diet and water for the full 10 h (control); (2) fasted broilers receiving only water (FW); (3) 1 g (0.01SW); (4) 2 g (0.02SW); and (5) 3 g (0.03SW) saccarose L^?1 supplemented water for the 10 h. The CIE colour values representing lightness (L*), redness (a*), and yellowness (b*) were measured on raw livers, breast and thigh meats. RESULTS: Full‐fed broilers had lower dressing percentage, higher abdominal fat and liver yield and lighter thigh meat coloration than the FW birds (P < 0.05). The 0.02SW and 0.03SW resulted in live shrinks that were lower (P < 0.05) than those found in FW broilers. The 0.02SW and 0.03SW resulted in liver colours that were lighter than those found in FW and 0.01SW birds but darker than from control birds (P < 0.05). CONCLUSION: It was concluded that saccarose supplementation during feed withdrawal appears to have a measurable impact on live shrink, and lightness of liver and thigh meat colour in broilers. Copyright © 2007 Society of Chemical Industry     

Performance, carcass, gastrointestinal tract and meat quality traits, and selected blood parameters of broilers fed diets supplemented with humic substances

BACKGROUND: The effects of humic substances (HS) on quality traits (colour, water‐holding capacity, and pH) of breast and thigh meats and on blood minerals (calcium and phosphorus) and on metabolites (glucose, total protein, triglyceride and cholesterols) as well as on the performance, carcass, and gastrointestinal tract (gut) traits of broilers were studied. In total, 480 male chicks (ROSS 308) were allocated to four treatments, each with five replicates containing 24 birds. While control birds were fed a diet (mash form) with no additives (0 HS), other treatment birds were fed with diets containing HS at 0.5 (0.5 HS), 1.0 (1 HS) or 1.5 (1.5 HS) g kg^?1 from 1 to 42 days of age. RESULTS: The body and carcass weights and feed efficiency increased (P < 0.05) by 1.5 HS, and blood cholesterol decreased (P < 0.05) by 1 HS and 1.5 HS compared to 0 HS. The 0.5 HS decreased (P < 0.10) low‐density lipoprotein compared to 0 HS. The 0.5 HS resulted in thigh meat and breast meat that were lighter and darker than those found in the other groups and 1.5 HS respectively (P < 0.01). CONCLUSION: Feeding with a diet containing HS caused a measurable variation in the meat quality and blood cholesterol as well as the performance, carcass, and gut traits of broilers. Copyright © 2011 Society of Chemical Industry     

Sensitivity study of the Drucker–Prager modeling parameters in the prediction of the nonlinear response of reinforced concrete structures

This study is aimed at investigating the sensitivity of the Drucker–Prager modeling parameters and the use of it in plasticity theory for shear design. The Drucker–Prager yield criterion is evaluated using finite element simulation with optimum mesh size for reinforced concrete (RC) beams with different geometrical and material properties. Providing an optimum mesh size configuration for the nonlinear analysis, the whole load–deflection behavior of the beams has been analyzed and the main results of the numerical analyses have been compared with the experimental results of RC beams to demonstrate how accurately the Drucker–Prager yield criterion for concrete predicts the load carrying capacity. The considered variables are compressive strength of concrete (f_c), tensile reinforcement ratio (ρ), shear span-to-depth ratio (a/d) and size effect (d/d_a). The nonlinear analyses of the beams, employing the Drucker–Prager yield criterion for concrete, have been performed and reasonable concrete cohesion has been investigated.     

The influence of dietary hazelnut kernel oil on the performance and fatty acid composition of broilers

A total of 360 one‐day‐old broilers (Ross 308) were allocated to three treatment groups with four replicates each of 30 chicks to evaluate the effects of the dietary hazelnut kernel oil (HKO) inclusion on the performance and the composition of fatty acid of meat in broilers. Treatments were diets with (1) soybean oil (SO), (2) HKO, or (3) HKO + SO (in equal amount, HKSO) added to the basal diets. There were no significant differences in weight gain, feed per gain, dressing percentage, abdominal fat weight, and mortality among the different dietary oil sources. Dietary HKO decreased (P < 0.01) linoleic and linolenic acids contents and the ratio of SFA:MUFA (saturated fatty acids:unsaturated fatty acids) of breast meats in broiler, and increased oleic acid content compared to SO. The result indicates that meat from broiler chickens fed diets containing HKO, singly or combined with SO could be an important source of oleic acid for human diets. Copyright © 2007 Society of Chemical Industry     

nOBS: an ns2 based simulation tool for performance evaluation of TCP traffic in OBS networks

Performance evaluation of tcp traffic in obs networks has been under intensive study, since tcp constitutes the majority of Internet traffic. As a reliable and publicly available simulator, ns2 has been widely used for studying tcp/ip networks; however ns2 lacks many of the components for simulating optical burst switching networks. In this paper, an ns2 based obs simulation tool (nobs), which is built for studying burst assembly, scheduling and contention resolution algorithms in obs networks is presented. The node and link objects in obs are extended in nobs for developing optical nodes and optical links. The ingress, core and egress node functionalities are combined into a common optical node architecture, which comprises agents responsible for burstification, routing and scheduling. The effects of burstification parameters, e.g., burstification timeout, burst size and number of burstification buffers per egress node, on tcp performance are investigated using nobs for different tcp versions and different network topologies.     

VoIP over DVB-RCS satellite systems: Trial results and the impact of adaptive speech coding using cross-layer design

VoIP is one of the most popular applications on the Internet. The DVB-S2 (digital video broadcasting-satellite version 2) and DVB-RCS (digital video broadcasting-return channel satellite) standards provide the potential for reliable and efficient voice and data transfer over satellite networks. This is important, as satellites can play a role in broadband service provision by addressing the cases of limited terrestrial access such as in rural or underserved areas. The first part of this paper presents the results of our VoIP trials with different commercial DVB-S/RCS satellite offers and popular internet telephony and videoconferencing applications (Skype, MSN, etc.). These results reveal that packet delay and jitter are strongly affected by the satellite network component as well as the type of speech codec used. Accordingly, research presented in the second part of this paper is focused on dynamic speech coding rate control adapted to the conditions of the underlying network, in which the satellite domain presents the most challenging portion of the end-to-end path. For this purpose, a novel cross-layer mechanism is proposed to facilitate and increase the accuracy of the speech coding rate adaptation mechanism. Cross-layer design is a relatively new idea aiming to exploit information exchange among layers of the protocol stack. Our simulation analyses show that the proposed cross-layer mechanism can help us in optimally adjusting the speech coding rate to maintain the user-perceived quality in terms of MOS (mean opinion score) in the face of time-varying available satellite channel capacity.