Gonzaga University
Complex and Adaptive Systems Group

 

Undergraduates using computational and mathematical techniques to investigate language, language processing, and other complex and adaptive systems.

My partners in this effort are Dr. Richard Cangelosi of the Department of Mathematics, Graham Moorehead of the Aon Corporation, and Dr. Mark Vandam of the Department of Speech and Hearing Sciences, Washington State University.  

We are grateful to the McDonald Work Award and the R.W. Gillette Award programs for generously and continuously supporting undergraduate research. 

IMG_5459.JPG

Meetings

Tuesdays 6 - 7

BCISE 006

Zoom Link for Anya

Research Assistants

Student Research Assistants

  • Hannah Noelle Vollan, Senior Computer Science

  • Leon Garcia-Camargo, Junior Computer Science

  • Jason Lunder, Junior, Computer Science

  • Anya Yeramilli, Senior, East Lake High School

  • Jesus Espinoza, Junior Computer Science

  • Claire Charvet, Freshman, Applied Mathematics

  • Grace Schaefer, Freshman, Chemistry

  • Avery Edmonds, Junior, Computer Science

  • Emily Bodenbender, Junior, Computer Science

 

Recent Alums

  • Phillip Fishburn, B.S.C.S, 2020

  • Jeb Kilfoyle, B.S. in Mathematics and B.S.C.S. 2019 & 2020.  Jeb is now a Ph.D. student in computer science, working on computational theory, University of New Mexico 

  • Jacob Krantz, B.S.C.S, 2018.  Jacob is a Ph.D. student in computer science, working on machine learning, Oregon State

  • Max Dulin, B.S.C.S, 2018.  Max is a Security Engineer at Security Innovation, Seattle

  • Luke Johnson, B.S.C.S, 2017.  Luke is an alum of the Gonzaga University Center for Genetic Algorithms, which preceded the current group.   He’s a Ph.D. student in computer science, working on cryptography, at the University of Connecticut.

Papers/Presentations

Professional Venues (bold indicates student co-authors)

De Palma, P., Garcia-Camargo, L., Kilfoyle, Jeb, VanDam, M. , Stover, J. (2021). Speech tested for Zipfian fit using rigorous statistical techniques. Proceedings of the Linguistic Society of America.

 

Krantz, J., Dulin, M., De Palma, P. (2019). Language-Agnostic Syllabification with Neural Sequence Labeling. 18th International Conference on Machine Learning and Applications (ICMLA), December 16-19, Boca Raton, Florida. 

 

Krantz, J., Dulin, M., De Palma, P., VanDam M. (2018). Syllabification by phone categorization.  GECCO 2018: The Genetic and Evolutionary Computation Conference, Kyoto, July 15th to 19th, 2018.  

Selected Student Venue Presentations


Cole, HJ. . (2022). Killer Whale Population Dynamics.  Presented at Spokane In7ercollegiate Research Conference, Whitworth University, April 27. 

Espinoza, J. (2022). Child-Directed Speech. Presented at Spokane In7ercollegiate Research Conference, Whitworth University, April 27. 

Garcia-Camargo, L. (2021).  Zipf's Law in Speech: Child Autism Spectrum Disorder.  Presented at Spokane Intercollegiate Research Conference, Gonzaga University, April 24. 

Olafson, C., Garcia-Camargo, L., Lunder, J. (2021). Machine Translation using Dependency Trees.  Presented at Spokane Intercollegiate Research Conference, Gonzaga University, April 24

Cole, H. (2021). Natural Language Processing for Interspecies Communication.

Presented at Spokane Intercollegiate Research Conference, Gonzaga University, April 24

Krantz, J., Dulin, M. (2019).  Language Agnostic Syllabification and Neural Networks.  B.S. Thesis presented at Spokane Intercollegiate Research Conference, Gonzaga University, April 26-27.. 

 

Krantz, J. Dulin, M. (2018). Syllabification by Categorization.  Spokane Intercollegiate Research Conference, Whitworth University, April 27-28.
 

Krantz, J., Dulin, M. (2017). Probabilistic Syllabification of English Words. Spokane Intercollegiate Research Conference, Gonzaga University, April 21-22. Best Oral Presentation.
 

Carter Timm, (2017). A Non-Uniform, Event-Driven Sampling Waveform Approximation Technique Applied to Context-Free Phone Classification for Automatic Speech Recognition. Spokane Intercollegiate Research Conference, Gonzaga University, April 21-22.
 

Birmingham, C. (2014).  Grammatical and Semantic Coherence as Related to N-Gram Size in the Brown Corpus. Spokane Intercollegiate Research Conference. Gonzaga University, April 26, 2014.

 

Haddock, Jamie (2011).  A Probabilistic Part of Speech Tagger.  Spokane Intercollegiate Research Conference, Whitworth University, April 16.

Research

Recent

  • A probabilistic, language-agnostic syllabifier

  • Do word frequencies in spontaneous speech form a power law distribution?
     

Current

  • Do word frequencies in the spontaneous speech of children diagnosed with autism spectrum disorder form a power law distribution? 

  • Classifying killer whale vocalizations.

  • In conjunction with Dr. Mark Vandam, Department of Speech and Hearing Sciences, Washington state University

    • Analysis of child-directed speech in day-long recordings of children and their families. 

    • Modeling the speech of galactosemic children 

  • Morpheein Enzyme Kinetics
    Burkholderia cenocepacia is an opportunistic pathogen particularly dangerous for cystic fibrosis patients.  It can cause a severe decline in lung function with the potential of developing into a life-threatening systemic infection.  Antibiotic resistance makes B. cenocepacia extremely difficult to treat.  The catalytic activity of enzyme 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-CoA) provides an energy source for the pathogen.  Therefore, understanding HMG-CoA presents a promising opportunity to treat infection by B. cenocepacia.  Our initial goal is to develop mathematical models of the morpheein kinetics exhibited by Burkholderia cenocepacia.

  • Orcinus orca Population Dynamics
    The orca or killer whale (Orcinus orca) is a toothed whale belonging to the oceanic dolphin family, of which it is the largest member.  Two distinct populations of Resident Killer Whales, known as the Northern and Southern Residents, occupy Canadian Pacific waters.  Northern Resident Killer Whales are listed as Threatened, and Southern Resident Killer Whales are listed as Endangered under the Species at Risk Act.  We seek to develop an agent-based model of the two Resident populations' dynamics to quantify the impact of food supply, toxic pollution, noisy waters, and physical disturbance that contribute to population wellbeing.

  • Translation using a tree-structured interlingua
    SoTA translation models perform a mapping from one sequence of tokens to another using "transformers", multi-headed self-attention networks, divided into encoders and decoders.  While transformers are good at associating different meanings of a gloss (the string of letters representing a word) with different vectors, they cannot reliably associate the same vector with the same meaning in contexts that don't change the meaning.  For this and other reasons it is reasonable to conclude that transformers are approaching an accuracy ceiling, and human language is above that ceiling.  This research addresses translation with a novel technique that may function in the same complexity class as human language.  As a free by-product, this technique would also solve some open NLU problems. 

  • Enabling privacy-protected research across medical data Outside of China, there is no existing straightforward way to allow medical research to benefit from the medical data currently esconced within the large EHR (electronic health record) systems in the US and other privacy-respecting nations.  Centralized systems exist (e.g. Cerner, Epic), but the access is extremely limited and cumbersome.  Such systems also have no plans to compensate the people from whom the data flow.  The impetus for this project begins with three values: 1. Privacy is paramount: your medical data never leaves your device  2. Access leads to innovation: If we enable any medical researcher in the world to ask any question, innovations will occur  3. Data is valuable: Whenever your data is used, you should be compensated.  Given these three values, the engineering questions become clear, but not easy.  This research will lead to an existence proof that such a technology is possible.  

  • SpreadLoss: investigation into the implications of the spline theory of neural networks.  Recent research from Randall Balestriero and Yan Lecun suggests a non-mainstream understanding of neural networks.  Instead of performing a high-dimensional diffeomorphism, NNs are better described as linear model lookups, or locality-sensitive hashing.  The case is clearest when using only rectilinear units on hidden layers.  This case will be explored.  If the spline theory is true, there could be interesting regularities in the spread of the average data batch across neuron-defined polyhedra.  This spread will be explored with and without a custom loss function, SpreadLoss, which will either encourage or discourage greater distribution of a batch across polyhedra.