Search Results (3)

Games provide an excellent testing ground for machine learning and artificial intelligence, offering diverse environments with strategic challenges and complex decision-making scenarios. This study seeks to design a self-learning artificial intelligent agent capable of playing the trick-taking stage of the popular card game Thousand, known for its complex bidding system and dynamic gameplay. Due to the game’s vast state space and strategic complexity, other artificial intelligence approaches, such as Monte Carlo Tree Search and Deep Counterfactual Regret Minimisation, are infeasible. To address these challenges, the enhanced version of the REINFORCE policy gradient algorithm is proposed. Introducing a score-related parameter $\beta$ designed to guide the learning process by prioritising valuable games, the proposed approach enhances policy updates and improves overall learning outcomes. Moreover, leveraging the off-policy experience replay, along with the importance weighting of behavioural policy, enhanced training stability and reduced model variance. The proposed algorithm was applied to the trick-taking stage of the popular game Thousand Schnapsen in a two-player setup. Four distinct neural network models were explored to evaluate the performance of the proposed approach. A custom test suite of selected deals and tournament evaluations was employed to assess effectiveness. Comparisons were made against two benchmark strategies: a random strategy agent and an alpha-beta pruning tree search with varying search depths. The proposed algorithm achieved win rates exceeding 65% against the random agent, nearly 60% against alpha-beta pruning at a search depth of 6, and 55% against alpha-beta pruning at the maximum possible depth. Full article

Fungal pathogens involved in grapevine trunk diseases (GTDs) may infect grapevines throughout their lifetime, from nursery to vineyard, via open wounds in stems, canes or roots. In vineyards, pruning wound protection products (PWPPs) offer the best means to reduce the chance of infection by GTD fungi. However, PWPPs may affect non-target microorganisms that comprise the natural endophytic mycobiome residing in treated canes, disrupting microbial homeostasis and indirectly influencing grapevine health. Using DNA metabarcoding, we characterized the endophytic mycobiome of one-year-old canes of cultivars Cabernet Sauvignon and Syrah in two vineyards in Portugal and Italy and assessed the impact of established and novel PWPPs on the fungal communities of treated canes. Our results reveal a large fungal diversity (176 taxa), and we report multiple genera never detected before in grapevine wood (e.g., Symmetrospora and Akenomyces). We found differences in mycobiome beta diversity when comparing vineyards (p = 0.01) but not cultivars (p > 0.05). When examining PWPP-treated canes, we detected cultivar- and vineyard-dependent alterations in both alpha and beta diversity. In addition, numerous fungal taxa were over- or under-represented when compared to control canes. Among them, Epicoccum sp., a beneficial genus with biological control potential, was negatively affected by selected PWPPs. This study demonstrates that PWPPs induce alterations in the fungal communities of grapevines, requiring an urgent evaluation of their direct and indirect effects on plants health with consideration of factors such as climatic conditions and yearly variations, in order to better advise viticulturists and policy makers. Full article

The minimax algorithm, also called the negamax algorithm, remains today the most widely used search technique for two-player perfect-information games. However, minimaxing has been shown to be susceptible to game tree pathology, a paradoxical situation in which the accuracy of the search can decrease as the height of the tree increases. Althöfer’s alternative minimax algorithm has been proven to be invulnerable to pathology. However, it has not been clear whether alpha-beta pruning, a crucial component of practical game programs, could be applied in the context of Alhöfer’s algorithm. In this brief paper, we show how alpha-beta pruning can be adapted to Althöfer’s algorithm. Full article