math-skill 1.0.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (55) hide show
  1. package/LICENSE +21 -0
  2. package/README.md +278 -0
  3. package/agents/math-critic.md +203 -0
  4. package/commands/abstraction.md +34 -0
  5. package/commands/algorithmic-thinking.md +34 -0
  6. package/commands/ask.md +21 -0
  7. package/commands/axiomatization.md +34 -0
  8. package/commands/causal-inference.md +34 -0
  9. package/commands/discrete-combinatorial.md +34 -0
  10. package/commands/game-theory.md +34 -0
  11. package/commands/induction-analogy.md +34 -0
  12. package/commands/information-theory.md +34 -0
  13. package/commands/logic-deduction.md +34 -0
  14. package/commands/modeling.md +37 -0
  15. package/commands/optimization.md +33 -0
  16. package/commands/probability-statistics.md +36 -0
  17. package/commands/symmetry-invariance.md +34 -0
  18. package/commands/topological-thinking.md +33 -0
  19. package/commands/transformation.md +33 -0
  20. package/docs/CLAUDE.md +187 -0
  21. package/docs/inspiration.md +113 -0
  22. package/knowledge-base/overview.md +230 -0
  23. package/package.json +60 -0
  24. package/skills/abstraction/SKILL.md +264 -0
  25. package/skills/abstraction/original-texts.md +175 -0
  26. package/skills/algorithmic-thinking/SKILL.md +371 -0
  27. package/skills/algorithmic-thinking/original-texts.md +256 -0
  28. package/skills/axiomatization/SKILL.md +213 -0
  29. package/skills/axiomatization/original-texts.md +160 -0
  30. package/skills/causal-inference/SKILL.md +374 -0
  31. package/skills/causal-inference/original-texts.md +167 -0
  32. package/skills/discrete-combinatorial/SKILL.md +286 -0
  33. package/skills/discrete-combinatorial/original-texts.md +185 -0
  34. package/skills/game-theory/SKILL.md +318 -0
  35. package/skills/game-theory/original-texts.md +131 -0
  36. package/skills/induction-analogy/SKILL.md +310 -0
  37. package/skills/induction-analogy/original-texts.md +154 -0
  38. package/skills/information-theory/SKILL.md +242 -0
  39. package/skills/information-theory/original-texts.md +127 -0
  40. package/skills/logic-deduction/SKILL.md +280 -0
  41. package/skills/logic-deduction/original-texts.md +173 -0
  42. package/skills/meta-selector/SKILL.md +188 -0
  43. package/skills/meta-selector/original-texts.md +105 -0
  44. package/skills/modeling/SKILL.md +318 -0
  45. package/skills/modeling/original-texts.md +165 -0
  46. package/skills/optimization/SKILL.md +292 -0
  47. package/skills/optimization/original-texts.md +168 -0
  48. package/skills/probability-statistics/SKILL.md +312 -0
  49. package/skills/probability-statistics/original-texts.md +193 -0
  50. package/skills/symmetry-invariance/SKILL.md +358 -0
  51. package/skills/symmetry-invariance/original-texts.md +221 -0
  52. package/skills/topological-thinking/SKILL.md +273 -0
  53. package/skills/topological-thinking/original-texts.md +164 -0
  54. package/skills/transformation/SKILL.md +264 -0
  55. package/skills/transformation/original-texts.md +216 -0
@@ -0,0 +1,273 @@
1
+ ---
2
+ name: topological-thinking
3
+ description: |
4
+ 科研模式触发:拓扑数据分析(TDA)、持续同调、同调计算、基本群与同调群分析、单纯复形与滤流构造。
5
+ 生活模式触发:理解连通性、鲁棒性分析、识别"什么在变化中持续存在"、判断表面不同的事物是否本质同类。
6
+ English Research trigger: topological data analysis (TDA), persistent homology, homology computation, fundamental group & homology group analysis, simplicial complex & filtration construction.
7
+ English Life trigger: understanding connectivity, robustness analysis, recognizing what persists through change, judging whether superficially different things are essentially the same kind.
8
+ ---
9
+
10
+ # 🌀 拓扑思想
11
+
12
+ > "可以拉伸弯曲但不能撕裂——真正重要的性质在连续变形中不变。"
13
+ > "You can stretch and bend but never tear — truly important properties survive continuous deformation."
14
+ >
15
+ > —— 拓扑学、同调理论、拓扑数据分析
16
+ > —— Topology, Homology Theory, Topological Data Analysis
17
+
18
+ ## 核心原则 / Core Principle
19
+
20
+ 拓扑学研究连续变形下保持不变的性质。甜甜圈与咖啡杯拓扑等价——它们都有一个洞。核心洞察:当精确距离不重要时,拓扑捕获本质结构——连通性、洞、维数。应用:数据分析(TDA)、网络连通性、鲁棒性分析。
21
+
22
+ Topology studies properties preserved under continuous deformation. A donut and coffee cup are topologically equivalent — both have one hole. Key insight: when exact distances don't matter, topology captures essential structure — connectedness, holes, dimension. Applications: data analysis (TDA), network connectivity, robustness analysis.
23
+
24
+ > **数学形式化**(科研模式参考 / Research Mode Reference)
25
+ >
26
+ > **同胚(homeomorphism)**:连续双射且逆也连续的映射 h: X → Y,是拓扑等价的标准。同伦(homotopy):连续变形过程本身连续,比同胚更粗的等价。微分同胚(diffeomorphism):光滑的双射且逆光滑,微分拓扑的等价标准。
27
+ >
28
+ > Homeomorphism: continuous bijection with continuous inverse h: X → Y, the standard of topological equivalence. Homotopy: the deformation process itself is continuous, a coarser equivalence. Diffeomorphism: smooth bijection with smooth inverse, the standard of differential topology.
29
+ >
30
+ > **欧拉示性数** χ = V - E + F:凸多面体 χ = 2,环面 χ = 0,配合可定向性是曲面的完全拓扑不变量。
31
+ >
32
+ > Euler characteristic χ = V - E + F: convex polyhedra χ = 2, torus χ = 0; a complete topological invariant for surfaces (paired with orientability).
33
+ >
34
+ > **基本群** π₁(X):基于基点的环路模去同伦变形;π₁(S¹) = Z(绕圈次数),π₁(S²) = 0(球面上所有环路可收缩),π₁(环面) = Z × Z(两个方向的绕圈)。
35
+ >
36
+ > Fundamental group π₁(X): loops based at a point modulo homotopic deformation; π₁(S¹) = Z (winding number), π₁(S²) = 0 (all loops contractible), π₁(torus) = Z × Z (winding in two directions).
37
+ >
38
+ > **同调群** H_n(X):n 维"洞"的计数:H₀ = 连通分量,H₁ = 1 维洞(环路),H₂ = 2 维空洞(围成的空腔)。
39
+ >
40
+ > Homology groups H_n(X): counting n-dimensional "holes": H₀ = connected components, H₁ = 1-holes (loops), H₂ = 2-dimensional voids.
41
+ >
42
+ > **Betti 数** β_n = rank H_n(X):n 维洞的数量:β₀ = 连通分量数,β₁ = 1 维洞数,β₂ = 2 维空洞数。
43
+ >
44
+ > Betti numbers β_n = rank H_n(X): number of n-dimensional holes: β₀ = components, β₁ = 1-holes, β₂ = 2-voids.
45
+ >
46
+ > **单纯复形(simplicial complex)**:由点、线段、三角形等单纯形组合而成的拓扑空间模型。
47
+ >
48
+ > Simplicial complex: a topological space model composed of simplices (points, edges, triangles, etc.).
49
+ >
50
+ > **滤流(filtration)**:从小到大改变邻域半径,在每个半径下构建单纯复形。
51
+ >
52
+ > Filtration: vary neighborhood radius from small to large, constructing simplicial complex at each scale.
53
+ >
54
+ > **持续同调(persistent homology)**:沿滤流计算同调群,长条对应真实拓扑特征,短条对应噪声。
55
+ >
56
+ > Persistent homology: compute homology groups along filtration; long bars = true topological features, short bars = noise.
57
+
58
+ > 详细数学依据见 `original-texts.md`
59
+
60
+ ## 不适用场景 / When NOT to Use
61
+
62
+ - **需要精确测量** `[科研/通用]`——拓扑只关心定性结构(连通、有洞),不提供距离、角度等精确度量 / When precise measurements matter—topology only addresses qualitative structure (connected, holes), not distances or angles
63
+ - **度量性质是核心** `[科研]`——拓扑等价的空间可以有完全不同的度量行为;度量空间的分析工具与拓扑不同 / When metric properties are essential—topologically equivalent spaces can have wildly different metric behavior; analysis requires more than topology
64
+ - **纯离散结构无连续性** `[科研]`——拓扑以连续性为基础,对无连续结构的纯离散问题不适用 / Purely discrete with no continuity—topology rests on continuity; inapplicable to purely discrete problems
65
+ - **需要精确数值决策** `[生活]`——当决策取决于"多少""多大"而非"是否连通""有没有洞"时,拓扑视角过于粗化 / When decisions depend on "how much" or "how big" rather than "is it connected" or "does it have holes"—topological perspective over-simplifies
66
+
67
+ ## 何时使用 / When to Use
68
+
69
+ ### 科研触发条件 / Research Triggers
70
+
71
+ - **需要定性分类**——不在乎"多大""多远",只在乎"是否连通""有多少洞" / Need qualitative classification—only care about connectedness and number of holes
72
+ - **形状比大小更重要**——研究对象的整体形态而非精确尺寸 / Shape matters more than size
73
+ - **需要鲁棒性分析**——拓扑不变量保证性质在连续扰动下不消失 / Need robustness analysis—invariants guarantee properties persist under continuous perturbation
74
+ - **数据具有标准统计无法捕获的形状**——聚类、空洞、环状结构 / Data has shape that standard statistics misses
75
+ - **需要变形下存活的不变量**——拓扑绝缘体、拓扑纠错 / Need invariant that survives deformation—topological insulators, topological error correction
76
+
77
+ ### 生活触发条件 / Life Triggers
78
+
79
+ - **需要理解连通性**——谁和谁连通、信息能否流通、关系网络是否完整 / Need to understand connectivity—who connects to whom, can information flow, is the network complete
80
+ - **需要识别"变化中的不变"**——不管表面怎么变,什么核心特征始终存在 / Need to identify "what stays the same through change"—what core features persist regardless of surface variation
81
+ - **需要判断本质同类**——看起来不同的东西,核心结构是否相同 / Need to judge essential sameness—do superficially different things share the same core structure
82
+ - **需要鲁棒性判断**——系统在扰动下是否稳定、哪些性质不会被消除 / Need robustness judgment—is the system stable under perturbation, which properties won't disappear
83
+ - **需要发现盲区**——哪里有信息无法到达的区域、哪里有断裂 / Need to discover blind spots—where information cannot reach, where there are gaps
84
+
85
+ ## 方法流程 / Method
86
+
87
+ ### 第一步:识别连续性要求 / Identify Continuity Requirements
88
+
89
+ **科研模式 / Research Mode**
90
+
91
+ 明确具体的等价标准:同胚(homeomorphism)、同伦(homotopy)、微分同胚(diffeomorphism)。确定什么可以改变(距离、角度、大小)什么必须固定(洞的数目、连通性、维数)。
92
+
93
+ Specify the exact equivalence standard: homeomorphism, homotopy, or diffeomorphism. Identify what can change (distances, angles, sizes) and what must stay fixed (number of holes, connectedness, dimension).
94
+
95
+ **生活模式 / Life Mode**
96
+
97
+ 什么可以改变,什么必须固定?——精确的数值不重要时,关注整体结构和连通性。问:哪些变化只是表面的(大小、外观、配置),哪些变化会破坏核心功能(连通断裂、关键环节消失)?
98
+
99
+ What can change, what must stay fixed? — When exact values don't matter, focus on overall structure and connectivity. Ask: which changes are merely surface-level (size, appearance, configuration), and which would break core function (connectivity severed, key links lost)?
100
+
101
+ **共通要点 / Shared Key Point**
102
+
103
+ 核心区分:可变(表面/度量)vs 不变(结构/拓扑)。/ Core distinction: variable (surface/metric) vs invariant (structure/topology).
104
+
105
+ ### 第二步:寻找拓扑不变量 / Find Topological Invariants
106
+
107
+ **科研模式 / Research Mode**
108
+
109
+ 核心不变量清单:
110
+ - **欧拉示性数** χ = V - E + F——凸多面体 χ = 2,环面 χ = 0,配合可定向性是曲面完全拓扑不变量
111
+ - **连通性**(path-connectedness)——从任意点能否沿连续路径到达任意其他点;连通分量数 = rank H₀
112
+ - **紧致性**(compactness)——每个开覆盖有有限子覆盖;Rⁿ 中紧致 ⇔ 有界 + 闭
113
+ - **基本群** π₁(X)——环路模去同伦变形;π₁(S¹) = Z, π₁(S²) = 0, π₁(环面) = Z × Z
114
+ - **同调群** H_n(X)——H₀ = 连通分量, H₁ = 1 维洞, H₂ = 2 维空洞
115
+ - **Betti 数** β_n = rank H_n——β₀ = 连通分量数, β₁ = 1 维洞数, β₂ = 2 维空洞数
116
+
117
+ Core invariant list:
118
+ - **Euler characteristic** χ = V - E + F — complete surface invariant with orientability
119
+ - **Connectedness** — number of components = rank H₀
120
+ - **Compactness** — Heine-Borel: compact in Rⁿ ⇔ bounded + closed
121
+ - **Fundamental group** π₁(X) — loops modulo homotopy; π₁(S¹) = Z, π₁(S²) = 0, π₁(torus) = Z × Z
122
+ - **Homology groups** H_n(X) — H₀ = components, H₁ = 1-holes, H₂ = 2-voids
123
+ - **Betti numbers** β_n = rank H_n — β₀ = components, β₁ = 1-holes, β₂ = 2-voids
124
+
125
+ **生活模式 / Life Mode**
126
+
127
+ 不管怎么变化,什么特征始终存在?——找到"不管表面怎么变,内核不变"的东西。比如:不管组织怎么改组,沟通渠道的连通性不变;不管市场怎么波动,供需的循环结构不变。
128
+
129
+ What features always exist regardless of change? — Find what stays the same at the core even as the surface changes. E.g.: no matter how an organization restructures, communication channel connectivity persists; no matter how markets fluctuate, the supply-demand loop structure persists.
130
+
131
+ **共通要点 / Shared Key Point**
132
+
133
+ 不变量 = 变化中的恒定。/ Invariants = constancy through change.
134
+
135
+ ### 第三步:用不变量分类 / Classify Using Invariants
136
+
137
+ **科研模式 / Research Mode**
138
+
139
+ 同欧拉示性数 → 同类型曲面;同基本群 → 同同伦类型空间;分类定理:紧致曲面由可定向性 + 欧拉示性数完全分类。
140
+
141
+ Same Euler characteristic → same surface type; same fundamental group → same homotopy type; classification theorem: compact surfaces classified by orientability + Euler characteristic.
142
+
143
+ **生活模式 / Life Mode**
144
+
145
+ 表面不同的东西核心结构是否相同?——甜甜圈和咖啡杯看起来不同,但核心结构一样(都有一个洞)。生活中看似不同的问题可能本质同类:两个部门的冲突模式可能是同一种循环依赖结构。
146
+
147
+ Do superficially different things share the same core structure? — A donut and coffee cup look different but share the same core structure (one hole). Seemingly different life problems may be essentially the same kind: conflict patterns in two departments may share the same circular dependency structure.
148
+
149
+ **共通要点 / Shared Key Point**
150
+
151
+ 分类靠不变量,不看表面看结构。/ Classification by invariants: look at structure, not surface.
152
+
153
+ ### 第四步:构造拓扑模型 / Construct Topological Model
154
+
155
+ **科研模式 / Research Mode**
156
+
157
+ **数据**:对点云数据构建滤流(filtration)——从小到大改变邻域半径,计算每个半径下的单纯复形及其同调;提取持续同调(persistent homology),输出条形码图(barcode diagram)或持续图(persistence diagram),长条 = 真实拓扑特征,短条 = 噪声。
158
+
159
+ **网络**:分析连通分量(可达性)、环路(冗余路径)、聚类系数;网络拓扑决定信息传播与鲁棒性。
160
+
161
+ **系统**:相空间中吸引子的拓扑——稳定不动点(拓扑上为点)、极限环(拓扑上为 S¹)、混沌吸引子(Strange attractor,分形结构)。
162
+
163
+ For data: persistent homology via filtrations—barcode/persistence diagram where long bars = true features, short bars = noise. For networks: connected components, cycles, clustering coefficient. For systems: attractor topology — fixed point, limit cycle (S¹), chaotic attractor (fractal).
164
+
165
+ **生活模式 / Life Mode**
166
+
167
+ 画出事物之间的连接关系——谁和谁连通,有没有断裂,有没有循环依赖,有没有空洞(信息无法到达的区域)。
168
+
169
+ Map the connections between things—who connects to whom, are there gaps, circular dependencies, voids (areas where information cannot reach).
170
+
171
+ **共通要点 / Shared Key Point**
172
+
173
+ 建模 = 把复杂现实映射到结构图。/ Modeling = mapping complex reality to a structural diagram.
174
+
175
+ ### 第五步:验证拓扑等价 / Verify Topological Equivalence
176
+
177
+ **科研模式 / Research Mode**
178
+
179
+ 尝试构造显式同胚或同伦等价;检查不变量是否匹配;若不变量不同 → 空间拓扑不同(可靠否定判断);若全部匹配 → 可能拓扑等价(但不能确定——不变量可能不完备)。
180
+
181
+ Attempt explicit homeomorphism or homotopy equivalence; check invariant matching; different invariants → topologically different (reliable negative); all match → possibly equivalent (invariants may be incomplete).
182
+
183
+ **生活模式 / Life Mode**
184
+
185
+ 验证你识别的核心结构是否真的可靠——变换是否真的不影响关键特征?问:如果表面变化了,我识别的"不变特征"还在吗?如果去掉一层包装,核心结构是否仍然成立?
186
+
187
+ Verify whether the core structure you identified is truly reliable — does transformation really not affect key features? Ask: if the surface changes, does the "invariant feature" remain? If you strip away packaging, does the core structure still hold?
188
+
189
+ **共通要点 / Shared Key Point**
190
+
191
+ 验证 = 确认不变量确实不变。/ Verification = confirming invariants are indeed invariant.
192
+
193
+ ### 第六步:拓扑推理与应用 / Topological Reasoning and Applications
194
+
195
+ **科研模式 / Research Mode**
196
+
197
+ 利用拓扑结论推理:
198
+ - "因为 π₁(X) ≠ 0,存在不可收缩环路 → 系统有不可达状态"
199
+ - "因为 X 连通,存在路径 → 转移总是可能的"
200
+ - "因为 β₁ = 3,数据中有 3 个显著的环状结构"
201
+ - "因为 χ = 0,这是环面类型的曲面"
202
+
203
+ Use topological conclusions for reasoning:
204
+ - "Since π₁(X) ≠ 0, non-contractible loops exist → system has inaccessible states"
205
+ - "Since X is connected, paths exist → transition is always possible"
206
+ - "Since β₁ = 3, data has 3 significant circular structures"
207
+ - "Since χ = 0, this is a torus-type surface"
208
+
209
+ **生活模式 / Life Mode**
210
+
211
+ 用结构洞察指导行动——连通性告诉你信息能否流通,空洞告诉你哪里有盲区,循环依赖告诉你哪里可能卡住。
212
+
213
+ Use structural insight to guide action—connectivity tells you whether information can flow, voids tell you where there are blind spots, circular dependencies tell you where things may get stuck.
214
+
215
+ **共通要点 / Shared Key Point**
216
+
217
+ 从结构到结论:不变量 → 性质 → 推理。/ From structure to conclusion: invariants → properties → reasoning.
218
+
219
+ ## 常见错误 / Common Errors
220
+
221
+ | 错误 / Error | 数学批评 / Mathematical Critique | 正确做法 / Correct Approach | 标签 / Tag |
222
+ |-------------|-------------------------------|---------------------------|-----------|
223
+ | 混淆拓扑等价与几何相似 / Confusing topological equivalence with geometric similarity | 拓扑等价只要求同胚,不要求度量相近;圆和椭圆拓扑等价但几何不同 | 区分"拓扑相同"和"几何相似";先问"连续变形能否到达?"再问"度量是否接近?" | `[科研/通用]` |
224
+ | 忽略维数差异 / Ignoring dimension differences | R¹ 与 R² 不同胚(维数是拓扑不变量——Brouwer 区域不变性定理) | 维数 invariance 是强制检查项;不同维数的空间拓扑必然不同 | `[科研]` |
225
+ | 混淆连通与道路连通 / Confusing connected with path-connected | 连通 ≠ 道路连通(如拓扑学家的正弦曲线);道路连通 ⇒ 连通,反之不成立 | 明确标注用的是哪种连通性;多数实际场景中道路连通更自然 | `[科研]` |
226
+ | 过度依赖单一不变量 / Over-relying on a single invariant | 仅用欧拉示性数无法区分所有空间(χ=0 的空间多种多样);单一不变量不完备 | 使用不变量组合(χ + 可定向性 + π₁ + H_n);不完备时标注不确定性 | `[科研/通用]` |
227
+ | 忽略局部与全局拓扑差异 / Ignoring local vs. global topological differences | 局部连通不保证全局连通;局部紧致 ≠ 紧致;局部同胚 ≠ 全局同胚 | 区分局部与全局性质;分别检查局部不变量与全局不变量 | `[科研]` |
228
+ | 混淆同调与同伦 / Confusing homology with homotopy | H₁(X) 是 π₁(X) 的 Abel 化(H₁ = π₁/commutators);同调是同伦的"粗化"版本 | 明确标注用的是 π₁(同伦)还是 H₁(同调);需要精细分类时用 π₁ | `[科研]` |
229
+ | 关注精确数值而忽略整体结构 / Focusing on exact values while ignoring overall structure | 拓扑的价值在于不看数值看结构;纠结精确数字会错失本质 | 先看连通性和循环结构,再看数值细节 / Look at connectivity and loop structure first, then numerical details | `[生活]` |
230
+ | 忽视断裂和空洞 / Ignoring gaps and voids | 断裂 = 连通性丧失,空洞 = 信息无法到达的区域;这些是最重要的拓扑特征 | 主动检查:有没有断裂?有没有信息盲区? / Proactively check: are there gaps? information blind spots? | `[生活]` |
231
+ | 把表面差异当本质差异 / Treating surface differences as essential differences | 甜甜圈和咖啡杯表面不同但本质相同;表面差异 ≠ 结构差异 | 问"去掉表面包装,核心结构是否同类?" / Ask: "stripping surface packaging, is the core structure the same kind?" | `[生活]` |
232
+
233
+ ## 操作规程 / Operating Procedure
234
+
235
+ 当本 skill 被触发时,首先选择模式:
236
+
237
+ **模式选择 / Mode Selection**
238
+ - **科研模式**:当涉及拓扑数据分析、持续同调、同调计算、单纯复形与滤流等数学/计算任务时
239
+ - **生活模式**:当涉及连通性理解、鲁棒性判断、识别变化中的不变、本质分类等日常推理任务时
240
+
241
+ **Research Mode**: for mathematical/computational tasks involving TDA, persistent homology, homology computation, simplicial complexes & filtrations
242
+ **Life Mode**: for everyday reasoning involving connectivity understanding, robustness judgment, identifying what persists through change, essential classification
243
+
244
+ ---
245
+
246
+ ### 科研模式输出格式 / Research Mode Output Format
247
+
248
+ 1. **连续性要求**:`[连续性要求]: [同胚/同伦/微分同胚],可变: [距离/角度/大小],不变: [洞数/连通性/维数]`
249
+ 2. **不变量**:`[不变量]: χ=[值] π₁=[群] β_n=[数], H₀=[连通分量数] H₁=[1维洞数] H₂=[2维空洞数],紧致性=[是/否]`
250
+ 3. **分类**:`[分类]: [可定向/不可定向], χ=[值] → [曲面类型];π₁=[群] → [同伦类型]`
251
+ 4. **拓扑模型**:`[拓扑模型]: [单纯复形/滤流/网络图/相空间],构造方法: [具体描述]`
252
+ 5. **等价验证**:`[等价验证]: [不变量是否全部匹配/是否有不变量不同],结论: [拓扑等价/拓扑不同/不确定]`
253
+ 6. **推理**:`[推理]: 因为[拓扑性质] → [结论]`
254
+
255
+ ### 生活模式输出格式 / Life Mode Output Format
256
+
257
+ 1. **[什么可变/什么不变]:[区分]** — 精确数值不重要,什么核心特征必须保持 / What can change vs. what must stay fixed — exact values don't matter, what core features must persist
258
+ 2. **[不变特征]:[发现]** — 不管怎么变化,什么始终存在 / Invariant features discovered — what always exists regardless of change
259
+ 3. **[本质分类]:[判断]** — 看似不同的东西核心是否同类 / Essential classification — are superficially different things the same kind at core
260
+ 4. **[连接结构]:[分析]** — 事物之间的连通、断裂、循环、空洞 / Connection structure analysis — connectivity, gaps, loops, voids between things
261
+ 5. **[结构可靠性]:[验证]** — 识别的结构是否真的可靠 / Structural reliability verification — is the identified structure truly reliable
262
+ 6. **[行动建议]:[步骤]** — 基于结构洞察的结论 / Action recommendation — conclusions based on structural insight
263
+
264
+ **输出必须包含以上 6 项,不得只输出分析性文字而不给出结论。**
265
+ **Output must include all 6 items above. Do not output only analytical text without conclusions.**
266
+
267
+ ## 与其他 skill 的关系 / Relations to Other Skills
268
+
269
+ - **对称与不变性**:拓扑不变量是连续变换下的不变量——拓扑思想与对称思想共享"寻找不变量"的核心逻辑,但拓扑关注连续变形而非群作用
270
+ - **抽象化思想**:拓扑是空间结构的抽象——去掉度量只保留开集与连续性,是从几何到拓扑的抽象跃升
271
+ - **变换思想**:连续变形是特殊的变换——同胚、同伦都是变换,拓扑研究这些变换下的不变量
272
+ - **建模思想**:拓扑数据分析是数据建模——用持续同调为点云数据建立拓扑模型,发现统计方法遗漏的形状特征
273
+ - **概率与统计**:TDA 补充统计的形状分析——统计关注均值与方差(量的分析),TDA 关注连通与空洞(质的分析),两者互补
@@ -0,0 +1,164 @@
1
+ # 数学出处与经典文献 / Mathematical Sources and Classic Texts
2
+
3
+ ## 欧拉示性数 / Euler Characteristic (Euler, 1752)
4
+
5
+ > "对于任何凸多面体,顶点数 - 边数 + 面数 = 2。"
6
+ > "For any convex polyhedron, the number of vertices minus edges plus faces equals 2."
7
+
8
+ **Leonhard Euler**(1707-1783)在 1752 年发现了多面体的这一不变量:
9
+
10
+ V - E + F = 2
11
+
12
+ 这一公式看似简单,实则深刻——它对所有凸多面体成立,无论多面体是立方体(8-12+6=2)、四面体(4-6+4=2)还是十二面体(20-30+12=2)。欧拉示性数是曲面的拓扑不变量:连续变形不改变 χ 的值。球面 χ = 2,环面 χ = 0,双环面 χ = -2——每增加一个洞,χ 减少 2。
13
+
14
+ **推广**:对任意曲面 S,χ(S) = 2 - 2g(可定向,亏格 g)或 χ(S) = 2 - g(不可定向,亏格 g)。欧拉示性数与可定向性共同构成紧致曲面的完全分类不变量。
15
+
16
+ This formula seems simple but is profound — it holds for all convex polyhedra, whether cube (8-12+6=2), tetrahedron (4-6+4=2), or dodecahedron (20-30+12=2). The Euler characteristic is a topological invariant of surfaces: continuous deformations do not change χ. Sphere χ = 2, torus χ = 0, double torus χ = -2 — each additional hole reduces χ by 2. Generalization: for any surface S, χ(S) = 2 - 2g (orientable, genus g) or χ(S) = 2 - g (non-orientable, genus g). Euler characteristic and orientability together form a complete set of classification invariants for compact surfaces.
17
+
18
+ ## Poincaré 猜想与 Perelman 证明 / Poincaré Conjecture & Perelman's Proof (2004)
19
+
20
+ > "每一个单连通的紧致三维流形都同胚于三维球面 S³。"
21
+ > "Every simply-connected closed 3-manifold is homeomorphic to the 3-sphere S³."
22
+
23
+ **Henri Poincaré**(1854-1912)在 1904 年提出这一猜想——它成为数学史上最著名的未解问题之一,困扰数学家整整一百年。
24
+
25
+ **Grigori Perelman**(1966-)在 2002-2003 年间在网上发布三篇论文,利用 **Ricci 流**(Hamilton 提出的微分几何工具)结合手术(surgery)技术完成了证明。Perelman 拒绝了 2006 年的 Fields 奖和 2010 年的 Clay 千禧奖百万美元奖金——数学史上最戏剧性的一幕。
26
+
27
+ **Ricci 流**的直觉:流形上的度量随时间演化,使得曲率趋于均匀——就像热传导使温度分布趋于均匀。当曲率集中在某处时("脖子"),用手术切除并补上标准球面,继续 Ricci 流。最终流形被分解为若干几何片(Thurston 的几何化猜想),每片都有标准度量——Poincaré 猜想是几何化猜想的特例。
28
+
29
+ Perelman's proof used **Ricci flow** (a tool introduced by Hamilton) combined with surgery. The intuition: the metric on a manifold evolves over time, making curvature tend toward uniformity — like heat diffusion making temperature uniform. When curvature concentrates at a "neck," surgery cuts it out and caps it with a standard sphere, then Ricci flow continues. The manifold is eventually decomposed into geometric pieces (Thurston's geometrization conjecture); Poincaré is a special case. Perelman declined the 2006 Fields Medal and 2010 Clay Millennium Prize — the most dramatic episode in mathematical history.
30
+
31
+ ## 基本群 / Fundamental Group (Poincaré, 1895)
32
+
33
+ > "基本群 π₁(X) 是所有基于基点的环路在同伦变形下的等价类——它捕获空间的'不可收缩环路'结构。"
34
+ > "The fundamental group π₁(X) consists of equivalence classes of loops based at a point modulo homotopic deformation — it captures the 'non-contractible loops' structure of a space."
35
+
36
+ **Poincaré** 在 1895 年的论文 *Analysis Situs* 中引入了基本群的概念——这是代数拓扑的奠基之作。
37
+
38
+ 核心例子:
39
+ - **π₁(S¹) = Z**——圆周上的环路按绕圈次数分类:绕 n 次的环路不能连续变形为绕 m 次的环路(n ≠ m)
40
+ - **π₁(S²) = 0**——球面上所有环路都可收缩为一点——球面"没有洞"
41
+ - **π₁(环面) = Z × Z**——两个独立方向的绕圈,组合为 (m, n) 绕圈次数
42
+ - **π₁(实射影平面 RP²) = Z₂**——只有两类环路:可收缩的(绕偶数次)和不可收缩的(绕奇数次)
43
+
44
+ **关键性质**:同胚空间有同构的基本群(拓扑不变量);但基本群不是完备不变量——不同空间可以有相同的基本群(如 S² 和 S³ 的 π₁ 都是 0,但它们不同胚)。
45
+
46
+ Core examples: π₁(S¹) = Z (winding number classifies loops); π₁(S²) = 0 (all loops contractible — sphere has "no holes"); π₁(torus) = Z × Z (winding in two independent directions); π₁(RP²) = Z₂ (even vs odd winding). Key property: homeomorphic spaces have isomorphic fundamental groups (topological invariant); but π₁ is not a complete invariant — different spaces can share the same π₁ (e.g., S² and S³ both have π₁ = 0 but are not homeomorphic).
47
+
48
+ ## 同调群 / Homology Groups
49
+
50
+ > "同调群 H_n(X) 计算空间中 n 维'洞'的数量——它将拓扑直觉转化为可计算的代数量。"
51
+ > "Homology groups H_n(X) count n-dimensional 'holes' in a space — translating topological intuition into computable algebraic quantities."
52
+
53
+ **单纯同调的计算流程**:
54
+ 1. 构造单纯复形 K(用三角形/四面体等单纯形近似空间)
55
+ 2. 定义边界算子 ∂_n: C_n → C_{n-1}(每个 n-单纯形的边界是 (n-1)-单纯形的组合)
56
+ 3. 关键性质 ∂² = 0(边界的边界为零)
57
+ 4. 同调群 H_n(K) = ker(∂_n) / im(∂_{n+1})——"环路"模去"边界"——真正的洞不是任何更高维单纯形的边界
58
+
59
+ **直觉**:H₀ = 连通分量数(零维"洞"= 分离的碎片),H₁ = 1 维洞数(不可填充的环路),H₂ = 2 维空洞数(不可填充的空腔)。Betti 数 β_n = rank H_n(X) 是洞的计数。
60
+
61
+ **同调 vs 同伦**:H₁(X) 是 π₁(X) 的 Abel 化(H₁ = π₁/[π₁, π₁],去掉交换子的约束)。同调计算更容易(线性代数),但信息更少(丢失了非交换结构);同伦信息更丰富但计算更难。选择取决于问题需求——需要精细分类用 π₁,需要快速计算用 H₁。
62
+
63
+ Simplicial homology computation: (1) build simplicial complex K; (2) define boundary operator ∂_n: C_n → C_{n-1}; (3) key property ∂² = 0; (4) H_n(K) = ker(∂_n) / im(∂_{n+1}) — "cycles" modulo "boundaries" — true holes are not boundaries of higher-dimensional simplices. Intuition: H₀ = connected components, H₁ = 1-holes (unfillable loops), H₂ = 2-voids (unfillable cavities). Betti numbers β_n = rank H_n. Homology vs homotopy: H₁(X) = π₁(X) abelianized. Homology is easier to compute (linear algebra) but less informative; homotopy is richer but harder. Choose based on needs — π₁ for fine classification, H₁ for fast computation.
64
+
65
+ ## 拓扑数据分析 / Topological Data Analysis (Carlsson, 2009)
66
+
67
+ > "持续同调让数据'说话'——它发现的不是均值和方差,而是数据的形状。"
68
+ > "Persistent homology lets data 'speak' — it discovers not means and variances, but the shape of data."
69
+
70
+ **Gunnar Carlsson**(Stanford)在 2009 年创立了 Ayasdi 公司,将持续同调推向应用。
71
+
72
+ **持续同调的计算流程**:
73
+ 1. 从点云数据出发,构造滤流(filtration):从小到大改变邻域半径 ε
74
+ 2. 对每个 ε,构建 Vietoris-Rips 复形 VR_ε(距离 ≤ ε 的点对连线,三点互距 ≤ ε 则填三角形)
75
+ 3. 对每个 ε 计算同调群 H_n(VR_ε)
76
+ 4. 记录每个拓扑特征的"出生时间"(ε_birth)和"死亡时间"(ε_death)
77
+ 5. 输出条形码图(barcode)或持续图(persistence diagram: 点 (ε_birth, ε_death))
78
+
79
+ **稳定性定理**:持续图之间的 Wasserstein 蕴距离 ≤ 数据之间的 Hausdorff 蕴距离——数据的微小扰动只导致持续图的微小变化。这保证了 TDA 的鲁棒性:噪声产生短条,真实特征产生长条。
80
+
81
+ **应用**:肿瘤基因数据的拓扑结构发现新的癌症亚型;传感器网络覆盖空洞检测;材料科学中相变拓扑标志;金融市场中市场崩溃前的拓扑预警。
82
+
83
+ Computation: (1) from point cloud, build filtration by varying neighborhood radius ε; (2) for each ε, construct Vietoris-Rips complex; (3) compute H_n at each ε; (4) record birth and death times; (5) output barcode or persistence diagram. Stability theorem: Wasserstein distance between diagrams ≤ Hausdorff distance between data — small data perturbation → small diagram perturbation. Noise produces short bars, real features produce long bars. Applications: tumor genomic topology revealing new cancer subtypes; sensor network coverage hole detection; phase transition signatures in materials science; topological early warning of market crashes in finance.
84
+
85
+ ## Brouwer 不动点定理 / Brouwer Fixed Point Theorem (1911)
86
+
87
+ > "每一个连续映射 f: Dⁿ → Dⁿ 都有不动点——即存在 x₀ 使得 f(x₀) = x₀。"
88
+ > "Every continuous map f: Dⁿ → Dⁿ has a fixed point — there exists x₀ with f(x₀) = x₀."
89
+
90
+ **L.E.J. Brouwer**(1881-1966)在 1911 年证明此定理——讽刺的是,Brouwer 后来成为直觉主义的创始人,反对他自己定理所依赖的排中律。
91
+
92
+ **直觉**:搅动杯中的咖啡,停搅后必有至少一点回到原位——因为搅动是圆盘上的连续映射,必有不动点。
93
+
94
+ **应用**:
95
+ - **博弈论**(Nash,1950):Nash 均衡的存在性依赖于 Brouwer 不动点定理——每个有限博弈至少存在一个混合策略 Nash 均衡
96
+ - **经济学**:一般均衡理论(Arrow-Debreu)的不动点证明——市场均衡的存在性
97
+ - **微分方程**:Peano 存在定理的证明用到不动点方法
98
+
99
+ Intuition: stir coffee in a cup, stop — some point must return to its original position, since stirring is a continuous map on a disk, which must have a fixed point. Applications: Nash equilibrium existence (Nash 1950) — every finite game has at least one mixed-strategy Nash equilibrium; economics — general equilibrium theory (Arrow-Debreu) uses fixed-point proof; differential equations — Peano existence theorem via fixed-point methods.
100
+
101
+ ## Jordan 曲线定理 / Jordan Curve Theorem
102
+
103
+ > "R² 中每条简单闭曲线将平面分成两个区域——内部和外部。"
104
+ > "Every simple closed curve in R² divides the plane into two regions — interior and exterior."
105
+
106
+ **看似显然却极难证明**——Jordan(1892)的原始证明有漏洞,Veblen(1905)给出第一个严格证明。定理的困难在于:闭曲线可以有极其复杂的拓扑行为(如 Osgood 曲线,面积可以任意大),直觉上的"显然"在数学上需要精细的拓扑论证。
107
+
108
+ **推广**:Jordan-Brouwer 分离定理——Rⁿ 中每个与 S^{n-1} 同胚的子集将 Rⁿ 分成两个连通区域。维数越高,证明越复杂。
109
+
110
+ Seemingly obvious yet extremely hard to prove — Jordan's original proof (1892) had gaps; Veblen (1905) gave the first rigorous proof. The difficulty: closed curves can have extremely complex topological behavior (e.g., Osgood curves with arbitrarily large area); what seems "obvious" intuitively requires fine topological argumentation mathematically. Generalization: Jordan-Brouwer separation theorem — every subset of Rⁿ homeomorphic to S^{n-1} divides Rⁿ into two connected components. Higher dimensions = more complex proofs.
111
+
112
+ ## 纽结理论 / Knot Theory
113
+
114
+ > "纽结的分类依靠不变量——Jones 多项式、Alexander 多项式等区分不同纽结类型。"
115
+ > "Knot classification relies on invariants — Jones polynomial, Alexander polynomial, etc. distinguish different knot types."
116
+
117
+ 纽结理论研究 R³ 中简单闭曲线的等价类(同痕 isotopy 下的分类)。核心问题:两个纽结能否通过连续变形(不切断不粘合)相互转化?
118
+
119
+ **关键不变量**:
120
+ - **Alexander 多项式**(1928):第一个纽结多项式不变量,但不完备(不同纽结可有相同 Alexander 多项式)
121
+ - **Jones 多项式**(1984):Vaughan Jones 的发现——与统计力学和量子场论有深层联系,Jones 因此获 1990 年 Fields 奖
122
+ - **纽结群** π₁(R³ - K):纽结在三维空间中的补空间的基本群
123
+
124
+ **应用**:DNA 拓扑——超螺旋(supercoiling)、纽结(knotted DNA)和链环(catenanes)影响 DNA 复制与转录;酶(拓扑异构酶 topoisomerase)改变 DNA 的拓扑结构,相当于进行"拓扑手术"。分子纽结化学——合成拓扑复杂的分子结构。
125
+
126
+ Knot theory studies equivalence classes of simple closed curves in R³ under isotopy. Key invariants: Alexander polynomial (1928, first polynomial invariant but incomplete); Jones polynomial (1984, deep connections to statistical mechanics and QFT, Jones won 1990 Fields Medal); knot group π₁(R³ - K). Applications: DNA topology — supercoiling, knotted DNA, and catenanes affect replication and transcription; topoisomerase enzymes perform "topological surgery" on DNA. Molecular knot chemistry — synthesizing topologically complex molecular structures.
127
+
128
+ ## 曲面分类定理 / Classification of Surfaces
129
+
130
+ > "紧致曲面由可定向性和欧拉示性数完全分类——这是拓扑分类的典范。"
131
+ > "Compact surfaces are completely classified by orientability and Euler characteristic — the paradigm of topological classification."
132
+
133
+ **分类定理**(Möbius 1861, Dyck 1888 完善):
134
+
135
+ 可定向曲面:S² (χ=2), 环面 T² (χ=0), 双环面 (χ=-2), 三环面 (χ=-4), ... — 亏格 g 的曲面 χ = 2 - 2g
136
+
137
+ 不可定向曲面:RP² (χ=1), Klein 瓶 (χ=0), ... — 亏格 g 的不可定向曲面 χ = 2 - g
138
+
139
+ **标准形式**:每个紧致曲面同胚于若干环面或若干 RP² 的连通和(connected sum)。可定向:#g T²(g 个环面的连通和);不可定向:#g RP²(g 个射影平面的连通和)。
140
+
141
+ **分类的完备性**:(可定向性, χ) 是完备不变量集——两个紧致曲面同胚 iff 它们有相同的可定向性和相同的 χ。
142
+
143
+ Classification theorem: orientable surfaces S² (χ=2), torus T² (χ=0), double torus (χ=-2), ... genus g surface χ = 2-2g. Non-orientable: RP² (χ=1), Klein bottle (χ=0), ... genus g non-orientable χ = 2-g. Standard form: every compact surface is homeomorphic to a connected sum of tori or projective planes. Completeness: (orientability, χ) is a complete invariant set — two compact surfaces are homeomorphic iff they share the same orientability and Euler characteristic.
144
+
145
+ ## Morse 理论 / Morse Theory
146
+
147
+ > "光滑函数的临界点编码了流形的拓扑——Morse 理论将分析与拓扑联系起来。"
148
+ > "Critical points of a smooth function encode the manifold's topology — Morse theory bridges analysis and topology."
149
+
150
+ **Marston Morse**(1892-1977)建立了这一理论:
151
+
152
+ **Morse 函数** f: M → R 的每个临界点 p(df(p) = 0)都是非退化的(Hessian 矩阵的行列式 ≠ 0)。临界点的指标(index)= Hessian 中负特征值的个数 = 临界点处"向下方向"的维数。
153
+
154
+ **核心定理**:Morse 不等式将临界点数与 Betti 数联系起来:
155
+ - c_k ≥ β_k(k 维临界点数 ≥ k 维 Betti 数)
156
+ - c_k - c_{k-1} + ... ± c_0 ≥ β_k - β_{k-1} + ... ± β_0
157
+
158
+ **直觉**:随着 f 的值从低到高扫描 M,每次经过一个指标为 k 的临界点,流形的拓扑发生一次"k 维附着"——相当于粘上一个 k 维胞腔。整个流形的拓扑由所有临界点及其指标完全决定。
159
+
160
+ Intuition: scanning M from low to high values of f, each critical point of index k triggers a "k-dimensional attachment" — gluing a k-cell. The manifold's entire topology is determined by all critical points and their indices. Core theorem: Morse inequalities relate critical point counts to Betti numbers — c_k ≥ β_k. Morse theory bridges differential analysis (critical points) and algebraic topology (homology).
161
+
162
+ ---
163
+
164
+ **总结 / Summary**:拓扑思想贯穿数学的核心——从 Euler 的示性数到 Poincaré 的基本群,从同调群的可计算性到持续同调的数据应用,从不动点定理的应用力量到纽结理论的生物关联,拓扑提供了"连续变形下什么不变"的统一视角。它告诉我们:真正重要的性质,不是精确的度量,而是结构的本质——连通、洞、维数、可定向性。这些性质在拉伸与弯曲中存活,在撕裂与粘合中改变——这正是拓扑区别于几何的根本界限。