Recognition of heterospecific mobbing calls can occur through both innate and learned mechanisms,with the former often explained by two main hypotheses:the acoustic similarity hypothesis,which emphasizes shared acoust...Recognition of heterospecific mobbing calls can occur through both innate and learned mechanisms,with the former often explained by two main hypotheses:the acoustic similarity hypothesis,which emphasizes shared acoustic features,and the phylogenetic conservatism hypothesis,which posits that closely related species may share innate decoding templates.However,it remains unclear whether phylogenetic relatedness alone can drive the recognition of unfamiliar mobbing calls,a question with important implications for understanding the evolution of interspecific communication and anti-predator strategies.We examined the recognition of unfamiliar mobbing calls in Masked Laughingthrushes(Pterorhinus perspicillatus) using playback experiments with three allopatric species' mobbing calls of Leiothrichidae family.Results revealed two key findings:(1) Masked Laughingthrushes exhibited mobbing responses to unfamiliar mobbing calls,though at significantly lower intensity compared to conspecific playbacks.(2) Phylogenetic relatedness significantly predicted mobbing intensity,independent of overall acoustic similarity.These findings improve our understanding of how birds like Masked Laughingthrush instinctively recognize mobbing calls from other species.We show phylogenetic relatedness rather than overall acoustic similarity may be a key to this innate ability.Species that share a common ancestor may possess similar built-in neural systems for decoding alarm signals.We suggest that future research needs to combine neurobiological techniques to determine how inherited biases and feature decoding system together guide variable bird communities to perceive heterospecific mobbing calls.展开更多
This paper introduces a quantum-enhanced edge computing framework that synergizes quantuminspired algorithms with advanced machine learning techniques to optimize real-time task offloading in edge computing environmen...This paper introduces a quantum-enhanced edge computing framework that synergizes quantuminspired algorithms with advanced machine learning techniques to optimize real-time task offloading in edge computing environments.This innovative approach not only significantly improves the system’s real-time responsiveness and resource utilization efficiency but also addresses critical challenges in Internet of Things(IoT)ecosystems—such as high demand variability,resource allocation uncertainties,and data privacy concerns—through practical solutions.Initially,the framework employs an adaptive adjustment mechanism to dynamically manage task and resource states,complemented by online learning models for precise predictive analytics.Secondly,it accelerates the search for optimal solutions using Grover’s algorithm while efficiently evaluating complex constraints through multi-controlled Toffoli gates,thereby markedly enhancing the practicality and robustness of the proposed solution.Furthermore,to bolster the system’s adaptability and response speed in dynamic environments,an efficientmonitoring mechanism and event-driven architecture are incorporated,ensuring timely responses to environmental changes and maintaining synchronization between internal and external systems.Experimental evaluations confirm that the proposed algorithm demonstrates superior performance in complex application scenarios,characterized by faster convergence,enhanced stability,and superior data privacy protection,alongside notable reductions in latency and optimized resource utilization.This research paves the way for transformative advancements in edge computing and IoT technologies,driving smart edge computing towards unprecedented levels of intelligence and automation.展开更多
文摘Recognition of heterospecific mobbing calls can occur through both innate and learned mechanisms,with the former often explained by two main hypotheses:the acoustic similarity hypothesis,which emphasizes shared acoustic features,and the phylogenetic conservatism hypothesis,which posits that closely related species may share innate decoding templates.However,it remains unclear whether phylogenetic relatedness alone can drive the recognition of unfamiliar mobbing calls,a question with important implications for understanding the evolution of interspecific communication and anti-predator strategies.We examined the recognition of unfamiliar mobbing calls in Masked Laughingthrushes(Pterorhinus perspicillatus) using playback experiments with three allopatric species' mobbing calls of Leiothrichidae family.Results revealed two key findings:(1) Masked Laughingthrushes exhibited mobbing responses to unfamiliar mobbing calls,though at significantly lower intensity compared to conspecific playbacks.(2) Phylogenetic relatedness significantly predicted mobbing intensity,independent of overall acoustic similarity.These findings improve our understanding of how birds like Masked Laughingthrush instinctively recognize mobbing calls from other species.We show phylogenetic relatedness rather than overall acoustic similarity may be a key to this innate ability.Species that share a common ancestor may possess similar built-in neural systems for decoding alarm signals.We suggest that future research needs to combine neurobiological techniques to determine how inherited biases and feature decoding system together guide variable bird communities to perceive heterospecific mobbing calls.
基金supported by National Natural Science Foundation of China(Nos.62071481 and 61501471).
文摘This paper introduces a quantum-enhanced edge computing framework that synergizes quantuminspired algorithms with advanced machine learning techniques to optimize real-time task offloading in edge computing environments.This innovative approach not only significantly improves the system’s real-time responsiveness and resource utilization efficiency but also addresses critical challenges in Internet of Things(IoT)ecosystems—such as high demand variability,resource allocation uncertainties,and data privacy concerns—through practical solutions.Initially,the framework employs an adaptive adjustment mechanism to dynamically manage task and resource states,complemented by online learning models for precise predictive analytics.Secondly,it accelerates the search for optimal solutions using Grover’s algorithm while efficiently evaluating complex constraints through multi-controlled Toffoli gates,thereby markedly enhancing the practicality and robustness of the proposed solution.Furthermore,to bolster the system’s adaptability and response speed in dynamic environments,an efficientmonitoring mechanism and event-driven architecture are incorporated,ensuring timely responses to environmental changes and maintaining synchronization between internal and external systems.Experimental evaluations confirm that the proposed algorithm demonstrates superior performance in complex application scenarios,characterized by faster convergence,enhanced stability,and superior data privacy protection,alongside notable reductions in latency and optimized resource utilization.This research paves the way for transformative advancements in edge computing and IoT technologies,driving smart edge computing towards unprecedented levels of intelligence and automation.
