Spatio-temporal credit assignment in neuronal population learning

In learning from trial and error, animals need to relate behavioral decisions to environmental reinforcement even though it may be difficult to assign credit to a particular decision when outcomes are uncertain or subject to delays. When considering the biophysical basis of learning, the credit-assignment problem is compounded because the behavioral decisions themselves result from the spatio-temporal aggregation of many synaptic releases. We present a model of plasticity induction for reinforcement learning in a population of leaky integrate and fire neurons which is based on a cascade of synaptic memory traces. Each synaptic cascade correlates presynaptic input first with postsynaptic events, next with the behavioral decisions and finally with external reinforcement. For operant conditioning, learning succeeds even when reinforcement is delivered with a delay so large that temporal contiguity between decision and pertinent reward is lost due to intervening decisions which are themselves subject to delayed reinforcement. This shows that the model provides a viable mechanism for temporal credit assignment. Further, learning speeds up with increasing population size, so the plasticity cascade simultaneously addresses the spatial problem of assigning credit to synapses in different population neurons. Simulations on other tasks, such as sequential decision making, serve to contrast the performance of the proposed scheme to that of temporal difference-based learning. We argue that, due to their comparative robustness, synaptic plasticity cascades are attractive basic models of reinforcement learning in the brain.     

Artificial metaplasticity neural network applied to credit scoring

The assessment of the risk of default on credit is important for financial institutions. Different Artificial Neural Networks (ANN) have been suggested to tackle the credit scoring problem, however, the obtained error rates are often high. In the search for the best ANN algorithm for credit scoring, this paper contributes with the application of an ANN Training Algorithm inspired by the neurons' biological property of metaplasticity. This algorithm is especially efficient when few patterns of a class are available, or when information inherent to low probability events is crucial for a successful application, as weight updating is overemphasized in the less frequent activations than in the more frequent ones. Two well-known and readily available such as: Australia and German data sets has been used to test the algorithm. The results obtained by AMMLP shown have been superior to state-of-the-art classification algorithms in credit scoring.     

Source of the soybean N credit in maize production

Nitrogen response trials throughout the United States Corn Belt show that economic optimum rates of N fertilization are usually less for maize (Zea mays L.) following soybean (Glycine max L.) than for maize following maize; however, the cause of this rotation effect is not fully understood. The objective of this study was to investigate the source of the apparent N contribution from soybean to maize (soybean N credit) by comparing soil N mineralization rates in field plots of unfertilized maize that had either nodulated soybean, non-nodulated soybean, or maize as the previous crop. Crop yields, plant N accumulation, soil inorganic N, and net soil mineralization were measured. Both grain yield (6.3 vs. 2.8 Mg ha^–1) and above-ground N accumulation (97 vs. 71 kg ha^–1) were greatly increased when maize followed nodulated soybean compared with maize following maize. A partial benefit to yield and N accumulation was also observed for maize following non-nodulated soybean. Cumulative net soil N mineralization following nodulated soybean, non-nodulated soybean, and maize was 112, 92 and 79 kg N ha^–1, respectively. Net mineralization of soil N appeared to be influenced by both quality (C:N ratio) and quantity of residue from the previous crop. In addition to an increase in plant available N from mineralization, the amount of soil inorganic N (especially in soil 5 cm from the row) was greater following nodulated soybean than non-nodulated soybean or maize. Based on these data, the soybean N credit appears to result from a combination of a decrease in net soil mineralization in continuous maize production and an increase in residual soil N from symbiotic fixation.     

Supply management of high-value components with a credit constraint

Due to rapid increases in the automotive markets in emerging economies, leading car manufacturers rapidly expand their manufacturing capacity in countries such as China. A certain portion of key performance components however continue to be sourced from developed countries. We study such a manufacturer that imports high-value components from a developed country. There are two available transportation modes: a slow mode with low cost and long and stochastic lead time, and a fast mode with high cost and short and deterministic lead time. Moreover, the manufacturer is subject to a credit constraint that bounds both the in-warehouse inventory and the number of outstanding orders (because it needs to pay for the components in advance). Consequently, the cheapest hedge against demand and supply uncertainty??inventory??is only available to a limited extent and the decision maker must turn to using the fast transportation mode for a much larger share of the orders. We model the manufacturer??s ordering policy and study its performance using simulation. Our study shows the adverse effects that the credit limit has on the growth opportunities of such companies in developing countries that import high-value components or other goods from developed countries. We show that especially the reduction in lead time variability can substantially reduce these adverse effects. Realizing that this variability is primarily caused by the import customs procedures, governments in developing economies have a means to assist their local manufacturers.