Category-Theory-for-AGI-UMass-CMPSCI-692CT

KO | EN

gitlite — search

#python #hacktoberfest #react #ai #golang #android #machine-learning #deep-learning #linux #docker #cli #pytorch #mcp #open-source

Category-Theory-for-AGI-UMass-CMPSCI-692CT

(★ 9)

Sample Google Colab code to illustrate application of category theory to AGI

Category-Theory-for-AGI-UMass-CMPSCI-692CT 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Category Theory for AGI — Colab Course Template

This bundle contains Colab-ready notebooks with a consistent **Environment** cell,
lightweight **micro-demos**, and **exercises**. Push this repo to GitHub and replace `sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT`
in each badge so students can launch directly in Colab.

The book now also has an official interface layer:

- `CLIFF` — the `Conscious Layer Interface to Functor Flow`

For a textbook-facing guide to using CLIFF together with these notebooks and
lecture materials, see:

- `CLIFF_APPENDIX.md`

## Structure
- `lectures/` (week-by-week lecture handouts/index pages)
- `notebooks/week01_backprop_as_functor.ipynb`
- `notebooks/week01_sudoku_gt_db.ipynb` (Sudoku GT vs Transformer + Diagrammatic Backprop)
- `notebooks/week01_lm_gt_vs_transformer.ipynb` (PTB language modeling: GT vs Transformer)
- `notebooks/week01_lm_causal_leakage_check.ipynb` (Causal leakage check: Transformer vs GT causal/non-causal)
- `notebooks/week01_coalgebra_demo.ipynb` (Coalgebra: deterministic + MDP)
- `notebooks/week02_yoneda_microdemo.ipynb`
- `notebooks/week02_yoneda_self_attention_demo1.ipynb` (Yoneda self-attention demo)
- `notebooks/week03_limits_colimits_microdemo.ipynb`
- `notebooks/week03_db_colimit_energy.ipynb` (DB as (co)limit energy minimization)
- `notebooks/week3_synthetic_av_pullback-2.ipynb` (Synthetic AV pullback alignment: hard vs soft pullback)
- `notebooks/week04_clustering_as_functor.ipynb`
- `notebooks/week04_gt_full_node_labeling.ipynb` (GT‑Full toy node labeling)
- `notebooks/week04_gt_causal_regimes_ptb_colab.ipynb` (PTB/WikiText-2 causal-ablation: GT causal/non-causal + transition variants)
- `notebooks/week05_mini_democritus.ipynb` (Mini‑Democritus causal manifold)
- `notebooks/week05_kan_extension_transformer_demo.ipynb` (Kan extension transformer demo)
- `notebooks/week05_attention_ket.ipynb` (Attention with KET components)
- `notebooks/week05_causal_attention_ket.ipynb` (Causal attention with KET components)
- `notebooks/week06_gt_full_ptb.ipynb` (GT‑Full on PTB, tiny LM)
- `notebooks/week07_sheaves_covers.ipynb` (Sheaves via covers and gluing)
- `notebooks/week08_topos_overlap_penalty.ipynb` (Topos overlap consistency penalty)
- `notebooks/week09_causal_discovery_toy.ipynb` (Toy causal discovery)
- `notebooks/week10_kan_do_rn.ipynb` (Kan-Do calculus with RN ratios)
- `notebooks/week11_subobject_classifier.ipynb` (Subobject classifier in Sets)
- `notebooks/week11_coalgebra_gtdb_rl_demo.ipynb` (Coalgebraic RL: GT vs GT+DB vs MLP)
- `notebooks/week12_jstability_regimes.ipynb` (j-stability across regimes)
- `notebooks/week13_democritus_manifold.ipynb` (Democritus causal manifold demo)
- `notebooks/week13_open_games_attention_economy.ipynb` (Open games: attention economy)
- `notebooks/Comparing_GT_vs_UMAP.ipynb` (GT+DB vs UMAP on relational domains beyond 1-skeleton constraints)
- `notebooks/Comparing_GT_vs_UMAP_DBLP.ipynb` (GT+DB vs UMAP on DBLP author-paper-term-conference relations)
- `data/democritus/manifold.npz` (small topic manifold snapshot)

Lecture index:

- `lectures/README.md`

## Usage
1. Open the notebook link (or upload to Colab), run the **Environment** cell first.
2. GPU is optional. All examples run on CPU in < 5 minutes.
3. For local use, create a conda env via `environment.yml` and run JupyterLab.

## CLIFF Companion

`CLIFF` is the official natural-language interface layer for *Categories for
AGI*.

It can help users:

- find the right chapter or concept
- choose the right notebook or demo
- connect the material to code snippets or project ideas
- move from textbook reading to runnable examples more quickly

Start with:

- `CLIFF_APPENDIX.md` for the book companion guide
- `CLIFF_CatAgi` for the released CLIFF interface repo

## Colab Badge Template
```
[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)]
(https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_backprop_as_functor.ipynb)
```
Replace `sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT` once you publish the repo.

## Local environment (optional)
```
conda env create -f environment.yml
conda activate ct4agi
jupyter lab
```

## Direct Colab Links

- Week 1 — Backprop as a Functor  
  Functorial view of backprop with a small, runnable micro-demo.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_backprop_as_functor.ipynb
- Week 1 — Sudoku GT + Diagrammatic Backprop  
  Compare **Geometric Transformer vs. baseline Transformer** on 4x4 Sudoku; optional DB curvature penalty.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_sudoku_gt_db.ipynb
- Week 1 — Language Modeling (PTB): GT vs Transformer  
  Compare **Geometric Transformer vs. baseline Transformer** on a tiny PTB language model.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_lm_gt_vs_transformer.ipynb
- Week 1 — Causal Leakage Check (Transformer vs GT)  
  Synthetic random-token next-token test to detect future-token leakage in non-causal geometric mixing.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_lm_causal_leakage_check.ipynb
- Week 1 — Coalgebra Demo  
  Deterministic system + MDP as coalgebras.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_coalgebra_demo.ipynb
- Week 2 — Yoneda Micro-Demo  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week02_yoneda_microdemo.ipynb
- Week 2 — Yoneda Self-Attention Demo 1  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week02_yoneda_self_attention_demo1.ipynb
- Week 3 — Limits & Colimits Micro-Demos  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week03_limits_colimits_microdemo.ipynb
- Week 3 — DB as (Co)limit Energy  
  Minimize consistency/gluing energy to approximate limits and colimits.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week03_db_colimit_energy.ipynb
  - Week 3 — Synthetic AV Pullback Alignment  
  Pullback as alignment: hard pullback (argmax buckets) vs soft pullback (Hungarian matching).  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week3_synthetic_av_pullback-2.ipynb
- Week 4 — Clustering as a Functor  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_clustering_as_functor.ipynb
- Week 4 — GT‑Full Toy Node Labeling  
  Relation‑aware GT‑Full message passing vs baseline MLP.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_gt_full_node_labeling.ipynb
- Week 4 — GT Causal Regimes on PTB  
  Causal-ablation comparison for six LM variants on PTB or WikiText-2; regime-aware interpretation and transition metric.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_gt_causal_regimes_ptb_colab.ipynb
- Week 5 — Mini‑Democritus  
  Causal triples → GT‑Full embeddings → 2D manifold.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_mini_democritus.ipynb
- Week 5 — Kan Extension Transformer Demo  
  Introductory KET model notebook for Kan extensions in sequence modeling.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_kan_extension_transformer_demo.ipynb
- Week 5 — Attention KET  
  Attention formulation with KET structure and comparisons.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_attention_ket.ipynb
- Week 5 — Causal Attention KET  
  Causal attention variant of the KET architecture.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_causal_attention_ket.ipynb
- Week 6 — GT‑Full on PTB (Tiny)  
  Relation‑aware GT‑Full LM vs baseline Transformer on PTB.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week06_gt_full_ptb.ipynb
- Week 7 — Sheaves via Covers and Gluing  
  Local covers + overlap consistency on a toy interference example.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week07_sheaves_covers.ipynb
- Week 8 — Topos Overlap Penalty  
  Sheaf-style overlap penalty prefers the true causal graph.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week08_topos_overlap_penalty.ipynb
- Week 9 — Toy Causal Discovery  
  Recover edges from a linear SEM with simple regressions.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week09_causal_discovery_toy.ipynb
- Week 10 — Kan-Do Calculus with RN Ratios  
  Conditioning vs intervention via Radon–Nikodym reweighting.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week10_kan_do_rn.ipynb
- Week 11 — Subobject Classifier  
  Characteristic maps in Sets as a concrete subobject classifier.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week11_subobject_classifier.ipynb
- Week 11 — Coalgebraic RL (GT + DB)  
  Harder synthetic MDP ablation: compare MLP, GT, and GT+DB under stochastic transitions.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week11_coalgebra_gtdb_rl_demo.ipynb
- Week 12 — j-Stability Across Regimes  
  Stable edges across environments from synthetic SEMs.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week12_jstability_regimes.ipynb
- Week 13 — Democritus Causal Manifold  
  Visualize a topic manifold as a lightweight DB + GT example.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week13_democritus_manifold.ipynb
- Week 13 — Open Games: Attention Economy  
  Simple open‑game play/co‑play demo for attention competition.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week13_open_games_attention_economy.ipynb
- Special Topic — GT+DB vs UMAP (Relational Domains)  
  Comparison notebook showing GT+DB embeddings on general relational data vs UMAP on 1-skeleton graph projections.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/Comparing_GT_vs_UMAP.ipynb
- Special Topic — GT+DB vs UMAP on DBLP  
  Real heterogeneous graph benchmark (author-paper-term-conference) contrasting 1-skeleton UMAP with triangle-aware GT.  
  https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/Comparing_GT_vs_UMAP_DBLP.ipynb

## Colab Badges

- Week 1 — Backprop as a Functor  
  Functorial view of backprop with a small, runnable micro-demo.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_backprop_as_functor.ipynb)
- Week 1 — Sudoku GT + Diagrammatic Backprop  
  Compare **Geometric Transformer vs. baseline Transformer** on 4x4 Sudoku; optional DB curvature penalty.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_sudoku_gt_db.ipynb)
- Week 1 — Language Modeling (PTB): GT vs Transformer  
  Compare **Geometric Transformer vs. baseline Transformer** on a tiny PTB language model.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_lm_gt_vs_transformer.ipynb)
- Week 1 — Causal Leakage Check (Transformer vs GT)  
  Synthetic random-token next-token test to detect future-token leakage in non-causal geometric mixing.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_lm_causal_leakage_check.ipynb)
- Week 1 — Coalgebra Demo  
  Deterministic system + MDP as coalgebras.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week01_coalgebra_demo.ipynb)
- Week 2 — Yoneda Micro-Demo  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week02_yoneda_microdemo.ipynb)
- Week 2 — Yoneda Self-Attention Demo 1  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week02_yoneda_self_attention_demo1.ipynb)
- Week 3 — Limits & Colimits Micro-Demos  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week03_limits_colimits_microdemo.ipynb)
- Week 3 — DB as (Co)limit Energy  
  Minimize consistency/gluing energy to approximate limits and colimits.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week03_db_colimit_energy.ipynb)
- Week 3 — Synthetic AV Pullback Alignment  
  Pullback as alignment: hard pullback (argmax buckets) vs soft pullback (Hungarian matching).  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week3_synthetic_av_pullback-2.ipynb)
- Week 4 — Clustering as a Functor  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_clustering_as_functor.ipynb)
- Week 4 — GT‑Full Toy Node Labeling  
  Relation‑aware GT‑Full message passing vs baseline MLP.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_gt_full_node_labeling.ipynb)
- Week 4 — GT Causal Regimes on PTB  
  Causal-ablation comparison for six LM variants on PTB or WikiText-2; regime-aware interpretation and transition metric.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week04_gt_causal_regimes_ptb_colab.ipynb)
- Week 5 — Mini‑Democritus  
  Causal triples → GT‑Full embeddings → 2D manifold.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_mini_democritus.ipynb)
- Week 5 — Kan Extension Transformer Demo  
  Introductory KET model notebook for Kan extensions in sequence modeling.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_kan_extension_transformer_demo.ipynb)
- Week 5 — Attention KET  
  Attention formulation with KET structure and comparisons.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_attention_ket.ipynb)
- Week 5 — Causal Attention KET  
  Causal attention variant of the KET architecture.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week05_causal_attention_ket.ipynb)
- Week 6 — GT‑Full on PTB (Tiny)  
  Relation‑aware GT‑Full LM vs baseline Transformer on PTB.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week06_gt_full_ptb.ipynb)
- Week 7 — Sheaves via Covers and Gluing  
  Local covers + overlap consistency on a toy interference example.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week07_sheaves_covers.ipynb)
- Week 8 — Topos Overlap Penalty  
  Sheaf-style overlap penalty prefers the true causal graph.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week08_topos_overlap_penalty.ipynb)
- Week 9 — Toy Causal Discovery  
  Recover edges from a linear SEM with simple regressions.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week09_causal_discovery_toy.ipynb)
- Week 10 — Kan-Do Calculus with RN Ratios  
  Conditioning vs intervention via Radon–Nikodym reweighting.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week10_kan_do_rn.ipynb)
- Week 11 — Subobject Classifier  
  Characteristic maps in Sets as a concrete subobject classifier.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week11_subobject_classifier.ipynb)
- Week 11 — Coalgebraic RL (GT + DB)  
  Harder synthetic MDP ablation: compare MLP, GT, and GT+DB under stochastic transitions.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week11_coalgebra_gtdb_rl_demo.ipynb)
- Week 12 — j-Stability Across Regimes  
  Stable edges across environments from synthetic SEMs.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week12_jstability_regimes.ipynb)
- Week 13 — Democritus Causal Manifold  
  Visualize a topic manifold as a lightweight DB + GT example.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week13_democritus_manifold.ipynb)
- Week 13 — Open Games: Attention Economy  
  Simple open‑game play/co‑play demo for attention competition.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/week13_open_games_attention_economy.ipynb)
- Special Topic — GT+DB vs UMAP (Relational Domains)  
  Comparison notebook showing GT+DB embeddings on general relational data vs UMAP on 1-skeleton graph projections.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/Comparing_GT_vs_UMAP.ipynb)
- Special Topic — GT+DB vs UMAP on DBLP  
  Real heterogeneous graph benchmark (author-paper-term-conference) contrasting 1-skeleton UMAP with triangle-aware GT.  
  [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/sridharmahadevan/Category-Theory-for-AGI-UMass-CMPSCI-692CT/blob/main/notebooks/Comparing_GT_vs_UMAP_DBLP.ipynb)

Category-Theory-for-AGI-UMass-CMPSCI-692CT

Category-Theory-for-AGI-UMass-CMPSCI-692CT 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Category-Theory-for-AGI-UMass-CMPSCI-692CT

Category-Theory-for-AGI-UMass-CMPSCI-692CT 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README