构建人工生命系统的设计考量与开放进化实验

需积分: 9 23 浏览量更新于2024-07-25 收藏 5.26MB PDF 举报

"《从人工进化到人工生命》是一本由Timothy John Taylor撰写的专业书籍，共317页，针对的是计算机科学与生物学交叉领域的核心议题——如何设计一个模拟生物进化过程的合成模型，即所谓的“人工生命”系统。该书试图探讨生物进化的基本逻辑结构，并着重关注开放性进化的理论和实践。在没有对生命确切定义的情况下，书中实验部分主要聚焦于具体问题，尤其是那些允许自我复制的计算机程序如何在模拟环境中进行并行处理和进化。作者介绍了一个名为Cosmos的人工进化系统，它是一个虚拟操作系统，能够仿真几千个自我复制程序的并行演化。Cosmos系统是对Ray的著名Tierra系统（Ray [Ray]）的扩展和改进，尽管它们之间存在联系，但Cosmos的设计和功能更为先进，它允许研究人员探索更深层次的生物进化机制和复杂性。这本书不仅提供了一个研究平台，还为理解生物进化的基本原理提供了新的视角。通过Cosmos，读者可以观察程序的适应性和创新性，以及它们如何通过自然选择和随机变异来发展。同时，它挑战了我们对于生命起源、复杂性的理解和控制可能性。这是一部深入探讨人工进化和人工生命设计策略的重要学术著作，对于计算机科学家、生物学家以及跨学科研究者来说，都具有很高的参考价值和启发意义。"

A.6 Implementation Details . . . . . . . . . . . . . . . . . . . . . . . . . . . 261

B Details of the Rep orted Cosmos Runs 263

B.1 Default Values for Minor Parameters in Cosmos . . . . . . . . . . . . . . 263

B.2 The Standard Ancestor Program, 348AAAA . . . . . . . . . . . . . . . . 264

B.3 The Sexual Ancestor Program, 1314AAAA . . . . . . . . . . . . . . . . 265

B.4 Genetic Co de . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269

B.5 RNG Seeds for Runs in Chapter 6 . . . . . . . . . . . . . . . . . . . . . 270

B.5.1 The Role of Chance . . . . . . . . . . . . . . . . . . . . . . . . . 270

B.5.2 Re-Running the Standard Mo del . . . . . . . . . . . . . . . . . . 271

B.5.3 Mutations and Flaws . . . . . . . . . . . . . . . . . . . . . . . . . 271

B.5.4 The CPU-time Distribution Scheme . . . . . . . . . . . . . . . . 271

B.5.5 Energy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

B.5.6 Reading Neighbouring Co de . . . . . . . . . . . . . . . . . . . . . 272

B.5.7 Ino culation with Sexual Ancestors . . . . . . . . . . . . . . . . . 272

Glossary 273

Bibliography 277

xvi

List of Figures

2.1 DierentTyp es of Complexity. . . . . . . . . . . . . . . . . . . . . . . . 28

4.1 The Structure of a Cell in Cosmos. . . . . . . . . . . . . . . . . . . . . . 76

4.2 Translation of the Genome. . . . . . . . . . . . . . . . . . . . . . . . . . 78

4.3 Searching for Communications with the

rms

receive

Instruction. . . . . 82

4.4 An Example of Organism Fission. . . . . . . . . . . . . . . . . . . . . . . 86

4.5 DierentPatterns of Energy Token Distribution. . . . . . . . . . . . . . 90

4.6 The Top-Level Algorithm. . . . . . . . . . . . . . . . . . . . . . . . . . . 98

5.1 Schematic of the Standard Ancestor, 348AAAA. . . . . . . . . . . . . . 112

5.2 Standard Run: Length. . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

5.3 Standard Run: Population Size. . . . . . . . . . . . . . . . . . . . . . . . 115

5.4 Standard Run: Replication Perio d. . . . . . . . . . . . . . . . . . . . . . 115

5.5 Standard Run: Diversity. . . . . . . . . . . . . . . . . . . . . . . . . . . 115

5.6 Standard Run: FlawPerio d. . . . . . . . . . . . . . . . . . . . . . . . . . 115

5.7 Standard Run: Age at Death. . . . . . . . . . . . . . . . . . . . . . . . . 115

5.8 Theoretical Relationship b etween Program Length and Replication Perio d.

Left

: Size of copy lo op increases prop ortionally with size of program.

Right

: Size of copy lo op is constant. . . . . . . . . . . . . . . . . . . . . 117

5.9 Population Size Limit Imp osed on Ancestor Programs by the Default

Rate of Intro duction of Energy into Environment. . . . . . . . . . . . . 120

5.10 Standard Run, Neutral Shadow: Diversity. . . . . . . . . . . . . . . . . . 121

5.11 Standard Run: Unfaithful Replications. . . . . . . . . . . . . . . . . . . 122

5.12 Standard Run: Unfaithful Replications (omitting p oints with zero in-

delity). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122

5.13 Standard Run: ActivityWaves Plot. . . . . . . . . . . . . . . . . . . . . 124

5.14 Standard Run, Neutral Shadow: ActivityWaves Plot. . . . . . . . . . . 124

xvii

5.15 Standard Run, Neutral Shadow: ActivityWaves Plot (magnied). . . . 124

5.16 Standard Run: Cumulative Activity. . . . . . . . . . . . . . . . . . . . . 125

5.17 Standard Run, Neutral Shadow: Cumulative Activity. . . . . . . . . . . 125

5.18 Standard Run: Mean Cumulative Activity. . . . . . . . . . . . . . . . . 125

5.19 Standard Run, Neutral Shadow: Mean Cumulative Activity. . . . . . . . 125

5.20 Standard Run: New Activity. . . . . . . . . . . . . . . . . . . . . . . . . 127

5.21 Standard Run, Neutral Shadow: New Activity. . . . . . . . . . . . . . . 127

5.22 Standard Run: Mean New Activity. . . . . . . . . . . . . . . . . . . . . 127

5.23 Standard Run, Neutral Shadow: Mean New Activity. . . . . . . . . . . . 127

5.24 Standard Run: Phylogeny of Signicant Genotypes. . . . . . . . . . . . 128

5.25 Standard Run: Cell Energy Distribution, Time Slice 5,000. . . . . . . . 131

5.26 Standard Run: Cell Energy Distribution, Time Slice 910,000. . . . . . . 131

5.27 Standard Run: Cell Age Distribution, Time Slice 900,000. . . . . . . . . 131

6.1 Contingency Exp eriments. Graphs of Cumulative Activity, Mean Cu-

mulative Activity, Diversity, Program Length and Replication Perio d

for Runs 10 (

left

) and 17 (

right

). . . . . . . . . . . . . . . . . . . . . . . 136

6.2 Pro cedure for Signicance Testing in Contingency Exp eriments. . . . . . 139

6.3 Cumulative Activity (concentration): Pairwise comparisons (

values)

between runs. Raw Sample Data (

left

). Dierenced Sample Data (

right

). 141

6.4 Mean Cumulative Activity (concentration): Pairwise comparisons between

runs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

6.5 Diversity: Pairwise comparisons b etween runs. . . . . . . . . . . . . . . 142

6.6 Program Length: Pairwise comparisons b etween runs. . . . . . . . . . . 142

6.7 Replication Period: Pairwise comparisons b etween runs. . . . . . . . . . 142

6.8

ActivityWave Diagram, Runs 10 (

left

) and 17 (

right

. . . . . . . . . . . . . 143

6.9 Class 2 versus Class 1 Dynamics. . . . . . . . . . . . . . . . . . . . . . . 147

6.10 Standard Mo del Re-Runs: Spatial distribution of programs in Run 1

during Class 1 phase. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

6.11 Standard Mo del Re-Runs: Age at Death of programs in Run 1. . . . . . 150

6.12 Standard Mo del Re-Runs: Number of faithful ospring of programs in

Run 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151

6.13 Population size graphs for ancestors with dierentnumb ers of

collect

instructions within copy lo op. . . . . . . . . . . . . . . . . . . . . . . . . 153

xviii

6.14 Low Mutation: Population size graphs for Run 3 (

left

) and Run 9 (

right

). 154

6.15 Low Mutation: Takeover of p opulation by `cancerous' mutant (darker

squares) in Run 3. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155

6.16 CPU-time Distribution: Graphs of Program Length for 4 of the 9 Runs. 157

6.17 Private Energy Collection: Class 1-like b ehaviour. . . . . . . . . . . . . 160

6.18 Private Energy Collection: Population Size and Diversity for Run 9. . . 161

6.19 Large Energy Gradient, Runs 1 and 2: Change in constitution of p opu-

lation over time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164

6.20 Large Energy Gradient, Runs 3, 5, 7 and 8: Change in constitution of

p opulation over time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165

6.21 Large Energy Gradient, Run 3: Spatial Distribution of Programs at

Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

6.22 Large Energy Gradient, Run 8: Spatial Distribution of Programs at

Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

6.23 Large Energy Gradient: Equilibrium distribution of environmental energy. 167

6.24 Large Energy Gradient, Run 5: Spatial Distribution of Programs of

Dierent Sizes at Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . 169

6.25 Large Energy Gradient, Run 7: Spatial Distribution of Programs of

Dierent Sizes at Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . 169

6.26 Large Energy Gradient, Runs 4, 6 and 9: Change in constitution of

p opulation over time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

6.27 Large Energy Gradient, Run 4: Spatial Distribution of Programs of

Dierent Sizes at Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . 173

6.28 Large Energy Gradient, Run 6: Spatial Distribution of Programs of

Dierent Sizes at Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . 173

6.29 Large Energy Gradient, Run 9: Spatial Distribution of Programs of

Dierent Sizes at Time Slice 900,000. . . . . . . . . . . . . . . . . . . . . 173

6.30 Schematic of the Sexual Ancestor, 1314AAAA. . . . . . . . . . . . . . . 178

7.1 Categorisation of Repro ducers. . . . . . . . . . . . . . . . . . . . . . . . 200

7.2 DNA Replication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210

7.3 Emb eddedness of Individuals and Richness of Interactions in Various

Articial Evolutionary Platforms. . . . . . . . . . . . . . . . . . . . . . . 217

A.1 Format of the

genetic

code.ini

le. . . . . . . . . . . . . . . . . . . . 255

A.2 Format of the

params.ini

le. . . . . . . . . . . . . . . . . . . . . . . . 255

A.3 Format of the

ancestor.ini

le. . . . . . . . . . . . . . . . . . . . . . . 256

xix

剩余316页未读，继续阅读

wqazwqaz

粉丝: 1
资源: 5

构建人工生命系统的设计考量与开放进化实验

人工智能-机器学习-试论人机关系的历史发展及其文化考量.pdf

Model Monitoring and Maintenance: 7 Key Steps to Ensure Long-Term Model Effectiveness

MATLAB Genetic Algorithm Case Studies: Combining Theory and Practice to Unveil the Essence of ...

macOS_Sequoia_15.1.password(imacos.top).rdr.split.016

【java毕业设计】小区物业管理系统（springboot+vue+mysql+说明文档）.zip

里面全部都是浪漫的爱心特效，有html和python编写的，大概几十种，欢迎下载，收藏

Delphi 12 控件之FUPX-32bit-PORTABLE.zip

HandyControl

macOS_Sequoia_15.1.password(imacos.top).rdr.split.049

“雅乐”私人牙科诊所管理系统的设计与实现ssm.zip

最新资源