My research focus centers on generative network models by combining methods in natural language processing and graph theory to learn the grammars of networks. In my free time you’ll find me fixing my house, running, sailing (less frequently nowadays), and spending time with friends, but most importantly with family, who are always reminding me to maintain a good sense of humor.
- CV
- Google Scholar Profile


Scheduled defense on Nov 2017 15th - Aguinaga, S., Generative Networks by Learning Hyperedge Replacement Grammars (PhD thesis).
Jun 2017 - Attending an presenting a poster at NetSci2017.
Jun 2017 - Growing Better Graphs with Latent-Variable Probabilistic Graph Grammars, Wang, et. al.; Paper under review.
Mar 2017 - Proposal funded: DOE Office of Science Graduate Student Research (SCGSR) Program award


HyperEdge Replacement Grammars Graph Model

Discovering the underlying structures present in large real-world graphs. Code

Automatically finding network architectures for Deep learning tasks

The difficulty of training deep neural networks (New).

Concept Hierarchies and Human Navigation

Solutions to the problem around the difficulty of training deep neural networks. Code, Paper


A mobile tool to automatically interact with family and close friends frienso.com Code.


A mobile tool to automatically monitor tremor over time (such as tremor due to Parkinson’s Disease) Paper

Get In Touch

Feel free to email me or contact me via social media or drop me a note.

  • Office:
    384K Nieuwland Science Hall
    Notre Dame, Indiana
    United States
  • 000-000-0000
  • saguinag@nd.edu