双相演化

此词条暂由彩云小译翻译，翻译字数共1046，未经人工整理和审校，带来阅读不便，请见谅。

}}

Dual phase evolution (DPE) is a process that drives self-organization within complex adaptive systems. It arises in response to phase changes within the network of connections formed by a system's components. DPE occurs in a wide range of physical, biological and social systems. Its applications to technology include methods for manufacturing novel materials and algorithms to solve complex problems in computation.

Introduction

Graphs and networks have two phases: disconnected (fragmented) and connected. In the connected phase every node is connected by an edge to at least one other node and for any pair of nodes, there is at least one path (sequence of edges) joining them.

图和网络有两个阶段: 不连接(支离破碎)和连接。在连接阶段，每个节点通过边连接到至少一个其他节点，对于任意一对节点，至少有一条路径(边序列)连接它们。

Dual phase evolution (DPE) is a process that promotes the emergence of large-scale order in complex systems. It occurs when a system repeatedly switches between various kinds of phases, and in each phase different processes act on the components or connections in the system. DPE arises because of a property of graphs and networks: the connectivity avalanche that occurs in graphs as the number of edges increases.

The Erdős–Rényi model shows that random graphs undergo a connectivity avalanche as the density of edges in a graph increases. In a local phase, the nodes behave as individuals; in the global phase, nodes are affected by interactions with other nodes. Most commonly the two processes at work can be interpreted as variation and selection. Variation refers to new features, which typically appear in one of the two phases. These features may be new nodes, new edges, or new properties of the nodes or edges. Selection here refers to ways in which the features are modified, refined, selected or removed. A simple example would be new edges being added at random in the global phase and edges being selectively removed in the local phase.

Erd s-Rényi 模型表明，随机图的边密度增加，图的连通度发生雪崩。在局部阶段，节点表现为个体; 在全局阶段，节点受到与其他节点交互的影响。最常见的两个过程在工作中可以解释为变异和选择。变异指的是新的特征，通常出现在两个阶段之一。这些特征可能是新的节点，新的边，或新的性质的节点或边。这里的选择指的是修改、细化、选择或删除特性的方法。一个简单的例子是在全局相位中随机添加新边，在局部相位中有选择地去除边。

Social networks provide a familiar example. In a social network the nodes of the network are people and the network connections (edges) are relationships or interactions between people. For any individual, social activity alternates between a local phase, in which they interact only with people they already know, and a global phase in which they can interact with a wide pool of people not previously known to them. Historically, these phases have been forced on people by constraints of time and space. People spend most of their time in a local phase and interact only with those immediately around them (family, neighbors, colleagues). However, intermittent activities such as parties, holidays, and conferences involve a shift into a global phase where they can interact with different people they do not know. Different processes dominate each phase. Essentially, people make new social links when in the global phase, and refine or break them (by ceasing contact) while in the local phase.

The DPE mechanism

The effects of changes in one phase carry over into the other phase. This means that the processes acting in each phase can modify or refine patterns formed in the other phase. For instance, in a social network, if a person makes new acquaintances during a global phase, then some of these new social connections might survive into the local phase to become long-term friends. In this way, DPE can create effects that may be impossible if both processes act at the same time.

一个阶段的变化的影响会传递到另一个阶段。这意味着在每个阶段中作用的过程可以修改或细化在另一个阶段中形成的模式。例如，在一个社交网络中，如果一个人在一个全球阶段结识了新朋友，那么这些新的社交关系中的一部分可能会在当地阶段存活下来，成为长期的朋友。通过这种方式，DPE 可以创建如果两个进程同时工作可能不可能的效果。

The following features are necessary for DPE to occur.

Underlying network

DPE has been found to occur in many natural and artificial systems.

DPE 已被发现存在于许多自然和人工系统中。

Small world networks, which are common in traditional societies, are a natural consequence of alternating local and global phases: new, long-distance links are formed during the global phase and existing links are reinforced (or removed) during the local phase. The advent of social media has decreased the constraining influence that space used to impose on social communication, so time has become the chief constraint for many people.

在传统社会中常见的小世界网络是地方和全球阶段交替的自然结果: 在全球阶段形成新的长途联系，在地方阶段加强(或删除)现有联系。社会媒体的出现减少了空间对社会交流的限制，因此时间成为许多人的主要限制。

DPE occurs where a system has an underlying network. That is, the system's components form a set of nodes and there are connections (edges) that join them. For example, a family tree is a network in which the nodes are people (with names) and the edges are relationships such as "mother of" or "married to". The nodes in the network can take physical form, such as atoms held together by atomic forces, or they may be dynamic states or conditions, such as positions on a chess board with moves by the players defining the edges.

The alternation between local and global phases in social networks occurs in many different guises. Some transitions between phases occur regularly, such as the daily cycle of people moving between home and work. This alternation can influence shifts in public opinion. Major fires (or other disturbances) clear away large tracts of land, so the network of free sites becomes connected and the landscape enters a global phase. In the global phase, competition for free sites is reduced, so the main competitive advantage is adaptation to the environment.

社会网络中局部和全局阶段之间的交替以许多不同的形式出现。有些阶段之间的转换是有规律的，例如人们每天在家和工作之间来回移动。这种变化会影响公众舆论的转变。大型火灾(或其他干扰)清除了大片土地，因此自由场地网络连接起来，景观进入全球阶段。在全球化阶段，免费网站的竞争减少，因此主要的竞争优势是适应环境。

In mathematical terms (graph theory), a graph [math]\displaystyle{ \textstyle G = \langle N,E\rangle }[/math] is a set of nodes [math]\displaystyle{ \textstyle N }[/math] and a set of edges [math]\displaystyle{ \textstyle E \subset \{ (x,y) \mid x,y \in N \} }[/math]. Each edge [math]\displaystyle{ \textstyle (x,y ) }[/math] provides a link between a pair of nodes [math]\displaystyle{ \textstyle x }[/math] and [math]\displaystyle{ \textstyle y }[/math]. A network is a graph in which values are assigned to the nodes and/or edges.

Most of the time a forest is in the local phase, as described above. The net effect is that established tree populations largely exclude invading species. A simple example is the Great Deluge algorithm in which the searcher can move at random across the landscape, but cannot enter low-lying areas that are flooded. At first the searcher can wander freely, but rising water levels eventually confine the search to a local area. Many other nature-inspired algorithms adopt similar approaches. Simulated annealing achieves a transition between phases via its cooling schedule. The cellular genetic algorithm places solutions in a pseudo landscape in which they breed only with local neighbours. Intermittent disasters clear patches, flipping the system into a global phase until gaps are filled again.

大多数时候，森林都处于局部阶段，如上所述。净效应是，已建立的树种群基本上排除了入侵物种。一个简单的例子是大洪水算法，在这种算法中，搜索者可以在地形上随机移动，但不能进入被淹没的低洼地区。起初，搜寻者可以自由漫步，但不断上升的水位最终将搜寻限制在局部地区。许多其他受自然启发的算法也采用了类似的方法。模拟退火通过其冷却时间表实现了两阶段之间的转换。细胞遗传算法将解决方案放置在一个只与当地邻居交配的伪景观中。断断续续的灾难清除了一小块一小块，将整个系统翻转到一个全球阶段，直到空隙再次被填满。

Phase shifts

Some variations on the memetic algorithm involve alternating between selection at different levels. These are related to the Baldwin effect, which arises when processes acting on phenotypes (e.g. learning) influence selection at the level of genotypes. In this sense, the Baldwin effect alternates between global search (genotypes) and local search (phenotypes).

模因算法的一些变化涉及不同层次的选择之间的交替。这些都与鲍德温效应有关，鲍德温效应是在表型作用过程中产生的。学习)影响基因型水平的选择。在这个意义上，Baldwin 效应在全局搜索(基因型)和局部搜索(表型)之间交替。

Graphs and networks have two phases: disconnected (fragmented) and connected. In the connected phase every node is connected by an edge to at least one other node and for any pair of nodes, there is at least one path (sequence of edges) joining them.

The Erdős–Rényi model shows that random graphs undergo a connectivity avalanche as the density of edges in a graph increases.<ref name="Erdos1960">

{{cite journal

Dual phase evolution is related to the well-known phenomenon of self-organized criticality (SOC). Both concern processes in which critical phase changes promote adaptation and organization within a system. However, SOC differs from DPE in several fundamental ways. Under SOC, a system's natural condition is to be in a critical state; in DPE a system's natural condition is a non-critical state. In SOC the size of disturbances follows a power law; in DPE disturbances are not necessarily distributed the same way. In SOC a system is not necessarily subject to other processes; in DPE different processes (e.g. selection and variation) operate in the two phases.

双相进化与众所周知的自组织临界性现象有关。两者都涉及关键阶段变化促进系统内部适应和组织的过程。然而，SOC 与 DPE 在几个基本方面有所不同。在 SOC 中，系统的自然状态是临界状态，在 DPE 中，系统的自然状态是非临界状态。在 SOC 中，扰动的大小遵循幂律，而在 DPE 中，扰动不一定按相同的方式分布。在 SOC 中，一个系统不一定受制于其他过程; 在 DPE 中，不同的过程(例如:。选择和变异)分两个阶段进行。

| author = Erdős, P.

| author2 = Rényi, A.

| name-list-style = amp

| year = 1960

| title = On the evolution of random graphs

Category:Nature-inspired metaheuristics

类别: 自然启发的启发式元推理

This page was moved from wikipedia:en:Dual-phase evolution. Its edit history can be viewed at 双相演化/edithistory