U.S.-Japan Seminar: Simulations of Complex Behavior from Simple Models

该奖项支持美国科学家参加将于2005年6月26日至29日在夏威夷坎纳帕利举行的美日科学家联合研讨会，研究用简单模型模拟复杂行为。协办人是雅典乔治亚大学的大卫·兰道教授和日本东京都大学的冈部裕高教授。统计物理学中的计算机模拟已经在物理学的各个领域取得了令人印象深刻的成果。随着并行计算平台的出现，模拟相当大的系统已经成为可能，我们现在可以认真考虑通过使用互补算法在长度尺度上跨越许多数量级的方法。在物理学中有许多问题，其中我们感兴趣的模型不是很复杂，但结果的行为是极其复杂的，并且进展需要特别的措施。关于某些模型是否具有向有序状态的相变的问题仍然存在。因此，研讨会的目标是重点了解在不同系统中出现的长时间尺度和复杂结构的问题，并研究克服这些困难的方法。感兴趣的主题范围从不同自旋模型中临界指数的确定到复杂系统（如玻璃）中相变的性质，以及复杂系统（如自旋玻璃和折叠蛋白质模型）中基态的描述。我们还将重点发展经典模型和量子模型的算法，以处理长时间尺度，包括在极低温度下发生的成核。共同组织者在该领域具有互补的科学专业知识。这使它们能够评估并吸引最优秀的科学研究人员参与。设计有效的策略，使这些物理系统的计算机模拟成为现实，可能会导致我们对蛋白质折叠、玻璃、生长和各种数学模型的研究取得重大进展，这些研究多年来一直面临挑战。这也很有可能刺激新算法的发展，刺激新科学的发展。研讨会的组织者已经做出了特别的努力，让博士后、研究生和未被充分代表的群体作为参与者和观察员。与该领域的日本专家交流想法和数据将使美国参与者能够推进自己的工作，并为未来的合作项目奠定基础。研讨会结束后，组织者计划建立一个研讨会网页。会议记录将在本网站公布。

0456399LandauThis award supports the participation of American scientists in a U.S.-Japan joint workshop for scientists on simulations of complex behavior from simple models to be held in Kannapali, Hawaii from June 26-29, 2005. The co-organizers are Professor David Landau at the University of Georgia in Athens and Professor Yutaka Okabe at the Tokyo Metropolitan University in Japan. Computer simulations in statistical physics have resulted in impressive gains in knowledge in diverse areas of physics. With the advent of parallel computing platforms, it has become possible to simulate quite large systems and we can now seriously consider ways to span many orders of magnitude in length scale by using complementary algorithms. There are many problems in physics where the model of interest is not very complicated but the resultant behavior is extremely complex and progress demands extraordinary measures. Questions still remain about whether or not certain models even have phase transitions to ordered states. Therefore, the goal of the workshop is to focus on understanding the problems with long time scales and complex structures which occur in diverse systems and to examine ways of overcoming these difficulties. The topics of interest range from the determination of critical exponents in different spin models to the nature of phase transitions in complex systems, such as glasses, and the description of groundstates in complex systems such as spin glasses and folded protein models. There will also be emphasis on development of algorithms for both classical and quantum models to deal with the long time scales including those which occur for nucleation at very low temperatures. The co-organizers have complementary scientific expertise in the field. This enables them to evaluate and attract the best science researchers to participate. Devising effective strategies that would make computer simulation of these physical systems practical could lead to significant advances in our study of protein folding, glasses, growth, and various mathematical models that have presented a challenge over the years. There is also a high probability of stimulating the development of new algorithms and stimulating new science. The workshop organizers have made a special effort to involve postdocs, graduate students and underrepresented groups as both participants and observers. The exchange of ideas and data with Japanese experts in the field will enable U.S. participants to advance their own work, and will set the stage for future collaborative projects. Following the seminar the organizers plan to establish a workshop web page. Conference proceedings will be published on this web site.