CV
Education
If you're curious about my transition from evolutionary biology to computer science.
Auburn University, August 2018 to August 2021
Ph.D., Computer Science and Software Engineering: Deep Learning/Machine Learning
Minor: Mathematics
Dissertation: "Understanding and Improving Computer Vision and Spatiotemporal Deep Neural Networks" (PDF, Recording, Slides)
Advisor: Anh Nguyen
Committee: Jan Van Haaren, Hans-Werner Van Wyk, and Levent Yilmaz
University Reader: Roberto Molinari
The University of Texas at Dallas, August 2013 to May 2015
M.S., Computer Science: Intelligent Systems
The University of Chicago, September 2010 to March 2013
M.S., Biology: Organismal Biology and Anatomy
Research Proposal: "The Anuran Pelvis: An Ecomorphological Analysis" (PDF)
Advisor: Mark Westneat
Auburn University, August 2006 to May 2010
B.S., Zoology: Ecology, Evolution, & Behavior (magna cum laude)
Thesis: "Head Shape and Intrasexual Behaviors of the Salamanders Eurycea aquatica and Eurycea cirrigera (Plethodontidae)" (PDF)
Advisor: Craig Guyer
MANUSCRIPTs
12. Alcorn, M. A. and N. Schwartz. 2024. Paved2Paradise: Cost-effective and scalable LiDAR simulation by factoring the real world. CVPR Workshop on Synthetic Data for Computer Vision. (PDF)
11. Alcorn, M. A. 2023. AQuaMaM: An autoregressive, quaternion manifold model for rapidly estimating complex SO(3) distributions. arXiv. (PDF)
10. Alcorn, M. A. and A. Nguyen. 2021. The DEformer: An order-agnostic distribution estimating Transformer. ICML Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models (INNF+). (PDF)
9. Alcorn, M. A. and A. Nguyen. 2021. baller2vec++: A look-ahead multi-entity Transformer for modeling coordinated agents. arXiv. (PDF)
8. Alcorn, M. A. and A. Nguyen. 2021. baller2vec: A multi-entity Transformer for multi-agent spatiotemporal modeling. arXiv. (PDF)
7. Li, Q., L. Mai, M. A. Alcorn, and A. Nguyen. 2020. A cost-effective method for improving and re-purposing large, pre-trained GANs by fine-tuning their class-embeddings. Asian Conference on Computer Vision (ACCV). (PDF)
6. Alcorn, M. A., Q. Li, Z. Gong, C. Wang, L. Mai, W. Ku, and A. Nguyen. 2019. Strike (with) a pose: Neural networks are easily fooled by strange poses of familiar objects. Conference on Computer Vision and Pattern Recognition (CVPR). (PDF) [Gizmodo, Gizmodo Australia, Nature, Nautilus, New Scientist, Rebooting AI, ZDNet]
5. Alcorn, M. A. 2018. (batter|pitcher)2vec: Statistic-free talent modeling with neural player embeddings. MIT Sloan Sports Analytics Conference. (PDF)
4. Deitloff, J., M. A. Alcorn, and S. P. Graham. 2014. Variation in mating systems of salamanders: mate guarding or territoriality? Behavioural Processes 106:111-116. (PDF)
3. Alcorn, M. A., J. Deitloff, S. P. Graham, and E. K. Timpe. 2013. Sexual dimorphism in head shape, relative head width, and body size of Eurycea aquatica and Eurycea cirrigera. Journal of Herpetology 47(2):321-327. (PDF)
2. Graham, S. P., M. A. Alcorn, E. K. Timpe, and J. Deitloff. 2013. Seasonal changes of primary and secondary sexual characteristics in the salamanders Eurycea aquatica and E. cirrigera. Herpetological Conservation and Biology 8(1):53-64. (PDF)
1. Graham, S. P., E. K. Timpe, S. K. Hoss, M. Alcorn, and J. Deitloff. 2010. Notes on reproduction in the brownback salamander (Eurycea aquatica). IRCF Reptiles & Amphibians 17(3):168-172. (PDF)
Presentations/Posters
28. Alcorn, M. A.. 2023. Logistic Regression in scikit-learn. Data Science Society at Auburn University (Invited). (Slides)
27. Alcorn, M. A., K. Geil, B. Stucky, and D. Peters. 2022. Modeling the Spread of a Livestock Disease With Semi-Supervised Spatiotemporal Deep Neural Networks. American Geophysical Union Fall Meeting. (Slides)
26. Ginn, D., P. J. Ramos-Giraldo, M. Alcorn, M. L. Cangiano, A. Dobbs, S. K. Skovsen, M. Kutugata, R. G. Leon, C. Reberg-Horton, M. Bagavathiannan, and S. B. Mirsky. 2022. 2022. Species-Level Weed Biomass Estimation from Video Imagery Using 3D Point Clouds. American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America (ASA/CSSA/SSSA) International Annual Meeting. (Poster) (Presented by D. Ginn)
25. Alcorn, M. A. 2022. Detecting GPS Errors in Livestock Collars with Factored Trajectory Transformers. Long-Term Agroecosystem Research Grazinglands Working Group Annual Meeting (Invited). (Slides)
24. Alcorn, M. A. 2022. Detecting GPS Errors in Livestock Collars with Factored Trajectory Transformers. SCINet Geospatial Workshop (Invited). (Slides)
23. Alcorn, M. A. 2022. A Shallow Introduction to Deep Learning. Charleston Data Science Meetup. (Slides)
22. Alcorn, M. A., B. Stucky, D. Peters, S. Mirsky, and C. Reberg-Horton. 2022. A Deep Learning Approach for Estimating the Impact of Cover Crops on Water Availability in Soil. Envisioning 2050 in the Southeast: AI-Driven Innovations in Agriculture. (Poster)
21. Alcorn, M. A. 2021. Deep Learning Approaches for Modeling Spatiotemporal Dynamics of a Livestock Disease. SCINet and AI COE Fellows Conference. (Slides)
20. Alcorn, M. A. 2021. A Shallow Introduction to Deep Learning. Data Science Society at Auburn University. (Slides)
19. Alcorn, M. A. and A. Nguyen. 2021. The DEformer: An Order-Agnostic Distribution Estimating Transformer. ICML Workshop on Invertible Neural Networks, Normalizing Flows, and Explicit Likelihood Models (INNF+). (Poster)
18. Alcorn, M. A. and A. Nguyen. 2021. baller2vec: A Multi-Entity Transformer For Multi-Agent Spatiotemporal Modeling. PyData Jeddah Meetup (Invited). (Recording, Slides)
17. Alcorn, M. A. and A. Nguyen. 2021. baller2vec: A Multi-Entity Transformer For Multi-Agent Spatiotemporal Modeling. Auburn Research Student Symposium. (Recording, Slides)
16. Alcorn, M. A. and A. Nguyen. 2021. baller2vec: A Multi-Entity Transformer For Multi-Agent Spatiotemporal Modeling. Auburn University – Statistics and Data Science Seminar (Invited). (Recording, Slides)
15. Alcorn, M. A. 2020. A Shallow Introduction to Deep Learning. Cleveland Indians Research & Development. (Slides)
14. Li, Q., L. Mai, M. A. Alcorn, and A. Nguyen. 2020. A Cost-Effective Method for Improving and Re-Purposing Large, Pre-Trained GANs by Fine-Tuning Their Class-Embeddings. Asian Conference on Computer Vision (ACCV). (Recording) (Presented by Q. Li)
13. Alcorn, M. A., Q. Li, Z. Gong, C. Wang, L. Mai, W. Ku, and A. Nguyen. 2019. Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects. Conference on Computer Vision and Pattern Recognition (CVPR). (Poster)
12. Alcorn, M. A., Q. Li, Z. Gong, C. Wang, L. Mai, W. Ku, and A. Nguyen. 2019. Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects. Auburn Research Student Symposium. (Slides)
11. Alcorn, M. A., Q. Li, Z. Gong, C. Wang, L. Mai, W. Ku, and A. Nguyen. 2019. Strike (with) a Pose: Neural Networks Are Easily Fooled by Strange Poses of Familiar Objects. Auburn University Council of Engineering Graduate Students – Finish in 5. (Slides)
10. Alcorn, M. A. 2018. (batter|pitcher)2vec: Statistic-Free Talent Modeling With Neural Player Embeddings. MIT Sloan Sports Analytics Conference. (Recording, Finals Recording, Slides, Poster)
9. Alcorn, M. A. 2017. Representation Learning @ Red Hat. EGG 2017 (Invited). (Slides)
8. Alcorn, M. A. 2017. Representation Learning @ Red Hat. MLconf San Francisco. (Recording, Slides)
7. Alcorn, M. A. 2016. More Than Words: Business Applications of Recurrent Neural Networks. PyData Carolinas. (Recording, Slides)
6. Alcorn, M. A. 2016. Deep Learning: The Open Source Way. Data4Decisions. (Slides)
5. Alcorn, M. A. 2010. Head Shape and Agonistic Behavior in the Salamanders Eurycea aquatica and E. cirrigera. National Conference on Undergraduate Research.
4. Alcorn, M. A. 2010. Head Shape and Intrasexual Behavior in the Salamanders Eurycea aquatica and E. cirrigera. Auburn University Undergraduate Research and Creative Scholarship Forum. (Slides)
3. Alcorn, M. A., J. Deitloff, and S. Graham. 2009. Sexual Dimorphism Differences in Eurycea aquatica and Eurycea cirrigera. Alabama Partners in Amphibian and Reptile Conservation. (Presented by J. Deitloff)
2. Alcorn, M. A., J. Deitloff, and S. Graham. 2009. Geometric Morphometrics Role in Phylogenetic Classification and Sexual Dimorphism Dynamics of Eurycea aquatica and Eurycea cirrigera. Joint Meeting of Ichthyologists and Herpetologists.
1. Alcorn, M. A., J. Deitloff, and S. Graham. 2009. Geometric Morphometrics’ Role in Phylogenetic Classification of Eurycea aquatica and Eurycea cirrigera. Association of Southeastern Biologists. (Poster)
Professional Experience
Senior CV/ML Engineer: February 2023 to present
• Role best summarized as "full stack deep learning scientist".
• Implemented, extended, trained, documented, and deployed (as containerized C++ ROS nodes) deep learning models for LiDAR and RGB perception that will eventually be used in safety critical environments.
• Invented an approach for cheaply generating diverse, realistic, and automatically annotated LiDAR scenes of humans in orchards.
• Defined protocols for data collection, data annotation, and model evaluation.
• Implemented highly optimized data processing pipelines for real-time perception on edge devices (e.g., NVIDIA Jetsons).
• Led paper discussions on state-of-the-art deep learning research.
• Technologies frequently used include: PyTorch, Open3D, NumPy, Numba, PyTorch3D, torchrun, Weights & Biases, Segments.ai, S3, git/GitLab, Torch-TensorRT, Docker, and ROS.
South Carolina Department of Natural Resources, Marine Resources Research Institute
Consultant: October 2022 to December 2022
• Implemented an agent-based model to simulate the population dynamics of a fish species and estimate the impact of different stocking strategies on genetic diversity.
United States Department of Agriculture Agricultural Research Service
Artificial Intelligence Center of Excellence Postdoctoral Fellow (ORISE): August 2021 to February 2023
• Worked under the mentorship of Brian Stucky and Deb Peters to develop deep learning models for predicting the spread of vesicular stomatitis in the United States.
• Collaborated with Steven Mirsky and Chris Reberg-Horton to develop deep learning models for estimating the impact of cover crops on soil water dynamics and implement computer vision models for detecting water stress in crops.
• Collaborated with Pat Clark to develop machine learning models for detecting errors in livestock GPS collars from tracking data.
• Implemented data acquisition pipelines pulling and linking data from disparate sources, including temporal geospatial data and satellite imagery.
• Assisted in workshops for training scientists in various technologies, including: Unix, Python, Git, and SQL.
Deep Learning Fellow: May 2020 to August 2020
• Worked with the Research & Development team to build deep learning models that informed game and season strategy.
Machine Learning Intern: May 2019 to August 2019
• Worked under the mentorship of Long Mai and Vova Kim, along with the Adobe Sensei team, to develop deep learning models for image retrieval and pose estimation.
Machine Learning Engineer – Information Retrieval: October 2017 to August 2018
Senior Software Engineer: September 2016 to September 2017
Software Engineer: June 2015 to August 2016
Software Engineering – Data Science Intern: May 2014 to May 2015
• Used a variety of machine learning, data science, and statistical techniques to address various company interests.
• Documented data collection, modeling decisions, and experimental results in internal blog posts and Jupyter notebooks.
• Introduced my team to state-of-the-art deep learning methods. Implemented various deep learning models in Keras, PyTorch, TensorFlow, Theano, and Lasagne, including long short-term memory (LSTM) recurrent neural networks (RNNs), which served as references for other employees.
• Spearheaded learning to rank efforts.
• Advocated for A/B testing and causal thinking.
• Prototyped a knowledge graph system using Semantic Web and information extraction technologies.
• Used a variety of natural language processing techniques to analyze text data.
• Co-founded and maintained Data Science @ Red Hat, a community of ~200 employees that hosted meetups and workshops, in addition to promoting knowledge sharing and collaboration.
• Provided technical guidance for and/or mentored coworkers, new hires, and interns, which involved: suggesting project ideas, suggesting relevant literature, providing example code, and providing technical feedback.
• Served as a liaison between different data groups at Red Hat, including Operations, Emerging Technologies, and Human Resources.
• Interviewed job candidates for data science positions, both on my team and other teams.
• Developed a business case and budget for upgraded machine learning hardware, which was approved.
• Regularly presented techniques and results to non-technical associates.
• Organized a weekly meeting between the data science interns where we discussed ideas and provided each other with feedback.
• Designed my team's logo.
Projects/Code
Paved2Paradise – cost-effective and scalable LiDAR simulation by factoring the real world.
PyTorch PointPillars – my minimal PyTorch implementation of the PointPillars model described here.
PyTorch PointNet++ – my minimal PyTorch implementation of the PointNet++ model described here.
AQuaMaM – an autoregressive, quaternion manifold model for rapidly estimating complex SO(3) distributions.
PyTorch implicit-PDF – my minimal PyTorch implementation of the implicit-PDF model described here.
PyTorch NeRF and pixelNeRF – my minimal PyTorch implementations of NeRF (as described here) and pixelNeRF (as described here).
PyTorch VQ-VAE – my minimal PyTorch implementation of the VQ-VAE model described here.
Boformer – a look-ahead multi-entity Transformer for jointly modeling coordinated agents in different sports.
DEformer – an order-agnostic distribution estimating Transformer.
PyTorch NADE – my PyTorch implementations of NADE (as described here) and order-agnostic NADE (as described here).
baller2vec++ – a look-ahead multi-entity Transformer for modeling coordinated agents.
baller2vec – a multi-entity Transformer for multi-agent spatiotemporal modeling.
PyTorch Geodesic Loss – implements a criterion for computing the distance between rotation matrices as described here and here.
PyTorch Volume Rotator – applies explicit 3D transformations to feature volumes in PyTorch.
BigGAN-AM – improves the sample diversity of BigGAN and synthesizes Places365 images using the BigGAN generator.
Strike (With) A Pose – a simple GUI application for generating adversarial poses of objects.
Apache Solr (contributor) – I implemented the RankNet scoring model as described here (Jira).
Learning To Name Colors With Word Embeddings – an improved version of the color name model described here.
LMIR – pure Python implementations of the language models for information retrieval surveyed here.
RankNet and LambdaRank – my (slightly modified) Keras implementation of RankNet (as described here) and PyTorch implementation of LambdaRank (as described here). My implementation was used as a reference by TripAdvisor in their photo ranking algorithm.
Recurrent Convolutional Neural Network Text Classifier – my (slightly modified) Keras implementation of the Recurrent Convolutional Neural Network (RCNN) described here.
scikit-learn (contributor) – I implemented the Complement Naive Bayes classifier described here (pull request).
Sequences With Sentences – a convolutional recurrent neural network that can handle data sequences containing a mixture of fixed size and variable size (e.g., text) inputs at each time step.
(batter|pitcher)2vec – a model inspired by word2vec that learns distributed representations of MLB players.
Deep Semantic Similarity Model – my Keras implementation of the Deep Semantic Similarity Model (DSSM)/Convolutional Latent Semantic Model (CLSM) described here.
LociSelect – a simulated annealing algorithm for selecting loci (locations in DNA) from the output of a high-throughput DNA sequencer for use in genetic analyses.
ARTificial Intelligence – a simple convolutional neural network that attempts to identify the movements and artists of visual art.
Hangouts NLP – a program that performs a number of different natural language processing analyses on instant messaging data.
Sexy Time Markov Model – a Markov model for simulating sex!
Football-o-Genetics – an application for "evolving" near-optimal offensive play calling strategies.
Social Networks in American Football – a social network analysis of coaches and teams in college and professional football.
ScatterPlot3D – an application for visualizing and exploring three-dimensional scatter plot data.
CyanogenMod – I developed an option that allows users to keep the status bar visible during "Expanded desktop" mode (pull requests: 1a and 1b).
Miscellaneous
My data science curriculum was featured by the GitHub Education Community.
"An Introduction to Machine-Learned Ranking in Apache Solr" – featured article on Opensource.com.
"Using Machine Learning to Name Colors" – featured article on Opensource.com.
"Coaching Football With AI" – featured article on Opensource.com.
"How to Become a Data Scientist" – featured article on Opensource.com.
"Representation Learning @ Red Hat" – featured blog post on MLconf.com (republished on Opensource.com).
Grants and Fellowships
2018-2021: Auburn University – Woltosz Fellow
2013-2014: The University of Texas at Dallas Jonsson School – Graduate Study Scholarship
2010-2012: The University of Chicago – Graduate Fellowship
2009-2010: Auburn University – Undergraduate Research Fellow
2009-2010: Auburn University Honors College – Drummond Company Scholar
2009: Fund for Excellence Undergraduate Research Award
2009: The Explorers Club Youth Activity Award
2006-2010: Auburn University – Academic Scholarship
2006-2010: College of Sciences and Mathematics Scholarship
Honors And Awards
2020: Huawei Best Application Paper Honorable Mention at the Asian Conference on Computer Vision (ACCV)
2019: Outstanding Graduate Student Council Senator
2019: Second Place Award for Oral Presentation in Science, Technology, Engineering and Mathematics at the Auburn Research Student Symposium ($250 prize)
2018: 3rd Place (out of "nearly 200") in the Research Papers Competition at the MIT Sloan Sports Analytics Conference ($1,000 prize)
2018: Opensource.com Emerging Contributor Award
2010: University Honors Scholar
2010: Phi Kappa Phi (invitations to top 5% of graduating class)
2010: College of Sciences and Mathematics – Dean’s Medal (Department of Biological Sciences)
2009: College of Sciences and Mathematics – Dean’s Research Award
2008: Auburn University Honors College – Junior Honors Certificate
2007-2009: Dean’s List
2007: COSAM Outstanding Freshman
2007: Alpha Lambda Delta (Honors Society)
2007: Phi Eta Sigma (Honors Society)
2005: Congressional Service Award
Research Positions
2014: Research Intern – Bioinformatics Core at the McDermott Center for Human Growth & Development, The University of Texas Southwestern Medical Center (Director: Chao Xing)
2009-2010: Research Assistant – Herpetological Collections, Auburn University Museum of Natural History
Teaching Experience
2022: Helper – Software Carpentry for the USDA (Instructors: James Smith, Thomas Guignard, and Steven Schroeder)
2022: Helper – Data Carpentry: Ecology with Python for the USDA (Instructors: Nishrin Kachwala, Mary-Francis LaPorte, and Jennifer Anne Wood Stubbs)
2021: Teaching Assistant – Intro to UNIX Command Line Webinar (Instructor: Brian Stucky)
2021: Teaching Assistant – Introduction to Computing Lab (Instructor: Hugh Kwon)
2021: Guest Lecturer – Wowapi Woecun Na Wounspe Wankatuyahci Glustanpi Kte Kin Hena (Research, Writing, and Statistics for Graduate Work) (Instructor: Richard Meyers)
2020: Teaching Assistant – Introduction to Computing Lab (Instructor: Hugh Kwon)
2018: Teaching Assistant – Fundamentals of Computing II Lab (Instructor: Dean Hendrix)
2018: Teaching Assistant – Ecology & Evolution in the Southwest Field Trip (Instructor: Eric Larsen)
2013: Instructor – Revolution Prep
2013: Substitute Teacher – Hurst-Euless-Bedford Independent School District
2012: Teaching Assistant – Tropical Ecology (Instructor: Eric Larsen)
2010: Peer Instructor – Conservation Biology Learning Community (Instructor: Robert Boyd)
Service
2022: Reviewer, Ecosphere
2021-2022: Reviewer, Machine Learning and Data Mining for Sports Analytics Workshop (MLSA)
2021: Reviewer, Artificial Intelligence in Sports Analytics Workshop (AISA)
2020: Volunteer (x3), Food Bank of East Alabama
2020: Reviewer, Springer
2019: Judge, Auburn Research Student Symposium
2019: Reviewer, Conference on Computer Vision and Pattern Recognition (CVPR)
2018-2019: Senator, Auburn University Graduate Student Council
2017-2018: Volunteer (weekly), RefugeeOne
2009-2010: Vice-President, Auburn University Chapter of the Society for Conservation Biology
2008-2009: Member, Auburn University Chapter of the Society for Conservation Biology
2005-2006: Founder and President, Hippies for Happiness (high school activism/charity organization)
2004-2006: Volunteer (weekly), St. Mary Hot Meal Program
2004-2005: Mentor, Big Brothers Big Sisters of America