arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1005.2714q-bio.PEcs.LG
收藏

结构漂移:序贯学习的种群动力学

Structural Drift: The Population Dynamics of Sequential Learning

Complexity Sciences Center(复杂性科学中心);University of California Davis(加州大学戴维斯分校);Santa Fe Institute(圣塔菲研究所);Columbia University(哥伦比亚大学)

James P. Crutchfield, Sean Whalen

摘要:

我们介绍了一个理论的顺序因果推理,在链中的学习者估计一个结构模型,从他们的上游教师,然后通过样本从模型到他们的下游学生。它扩展了遗传漂变的种群动力学,重铸木村的选择性中性学说〈动〉<动>作为一个特殊的情况下,广义漂移过程中使用结构化的人口与记忆。我们研究了几个漂移过程的扩散和固定特性,并提出了学习,推理和进化的应用。我们还展示了漂移过程空间的组织如何控制保真度,促进创新,并导致信息丢失的顺序学习和无记忆。

英文摘要:

We introduce a theory of sequential causal inference in which learners in a chain estimate a structural model from their upstream teacher and then pass samples from the model to their downstream student. It extends the population dynamics of genetic drift, recasting Kimura's selectively neutral theory as a special case of a generalized drift process using structured populations with memory. We examine the diffusion and fixation properties of several drift processes and propose applications to learning, inference, and evolution. We also demonstrate how the organization of drift process space controls fidelity, facilitates innovations, and leads to information loss in sequential learning with and without memory.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑