Fork me on GitHub

Education


University of Notre Dame — Notre Dame, IN

PhD Program in Computer Science & Engineering - Degree Awarded (May 2024)

GPA:

4.00 / 4.00

Rochester Institute of Technology (RIT) — Rochester, NY

BS / MS Program in Computer Engineering - All Degrees Awarded (May 2019)
magna cum laude

Grad GPA:

4.00 / 4.00

Undergrad GPA:

3.63 / 4.00

Cumulative GPA:

3.70 / 4.00

Courses
  • Brain-Inspired Computing
  • Computer Vision
  • Deep Learning
  • Advanced C++ Programming
  • Graph Theory
  • Computer Architecture
  • Applied Programming
  • Linear Optimization
  • Digital Systems I and II
  • Computer Science I and II
  • Electronics I

Publications


Also see my Google Scholar profile and ORCID Logo

Z. Carmichael, T. Redgrave, D. Gonzalez, W. J. Scheirer, "This Probably Looks Exactly Like That: An Invertible Prototypical Network," Proceedings of the European Conference on Computer Vision, Milan, Italy, 2024.
arXiv

Z. Carmichael, "Explainable AI for High-Stakes Decision-Making," Ph.D. Dissertation, Department of Computer Science and Engineering, University of Notre Dame, 2024.
Curate ND

Z. Carmichael, S. Lohit, A. Cheerian, M. Jones, W. J. Scheirer, "Pixel‐Grounded Prototypical Part Networks," Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, Hawaii, 2024.
WACV

S. Sarkar, A. Ramesh Babu, S. Mousavi, Z. Carmichael, V. Gundecha, S. Ghorbanpour, R. Luna Gutierrez, A. Guillen, A. Naug, "Benchmark Generation Framework with Custom Distortions for Image Classifier Robustness," Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, Waikoloa, Hawaii, 2024.
WACV

†§‡Z. Carmichael, W. J. Scheirer, "Unfooling Perturbation-Based Post Hoc Explainers," Proceedings of the AAAI Conference on Artificial Intelligence, Washington D.C., 2023.
AAAI

†§‡Z. Carmichael, T. Moon, S. A. Jacobs, "Learning Debuggable Models Through Multi-Objective Neural Architecture Search," International Conference on Automated Machine Learning (AutoML) Workshop, 2023.
AutoML

Z. Carmichael, W. J. Scheirer, "How Well Do Feature-Additive Explainers Explain Feature-Additive Predictors?," NeurIPS Workshop XAI in Action: Past, Present, and Future Applications, New Orleans, LA, 2023.
arXiv

S. J. Abraham, K. D. G. Maduranga, J. Kinnison, Z. Carmichael, J. D. Hauenstein, W. J. Scheirer, "HomOpt: A Homotopy‐Based Hyperparameter Optimization Method," Under Review, 2023.
arXiv

†§‡W. Theisen, D. Gonzalez, Z. Carmichael, T. Weninger, W. J. Scheirer, "Motif Mining: Finding and Summarizing Remixed Image Content," Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), Waikoloa, Hawaii, 2023.
WACV

†‡J. Takeshita, Z. Carmichael, R. Karl, T. Jung, "TERSE: Tiny Encryptions and Really Speedy Execution for Post-Quantum Private Stream Aggregation," Proceedings of the EAI International Conference on Security and Privacy in Communication Networks (SecureComm), Remote, 2022.
SecureComm

†§N. Soures, D. Chambers, Z. Carmichael, A. Daram, D. P. Shah, K. Clark, L. Potter, D. Kudithipudi, "SIRNet: Understanding Social Distancing Measures with Hybrid Neural Network Model for COVID-19 Infectious Spread," Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI) Disease Computational Modeling Workshop, Remote, 2020.
IJCAI

—, —, Proceedings of the International Conference on Machine Learning (ICML) Machine Learning for Global Health Workshop, Remote, 2020.
Poster

†§Z. Carmichael, H. F. Langroudi, C. Khazanov, J. Lillie, J. L. Gustafson, D. Kudithipudi, "Deep Positron: A Deep Neural Network Using Posit Number System," Proceedings of the IEEE Conference and Exhibition on Design, Automation and Test in Europe (DATE), Florence, Italy, 2019 [24% acceptance rate].
arXiv

†§Z. Carmichael, D. Kudithipudi. "Stochastic Tucker-Decomposed Recurrent Neural Networks for Forecasting." In IEEE Global Conference on Signal and Information Processing (GlobalSIP 2019), Ottawa, Canada, 2019.
GlobalSIP

†§Z. Carmichael, H. F. Langroudi, C. Khazanov, J. Lillie, J. L. Gustafson, D. Kudithipudi, "Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks," Proceedings of the ACM Conference for Next Generation Arithmetic (CoNGA), Singapore, 2019.
arXiv

†§Z. Carmichael, B. Glasstone, F. Cwitkowitz, K. Alexopoulos, R. Relyea, R. Ptucha, "Autonomous Navigation Using Localization Priors, Sensor Fusion, and Terrain Classification," Proceedings of IS&T International Symposium on Electronic Imaging: Image Processing Algorithms and Systems, San Francisco, CA, 2019.

†§Z. Carmichael, H. Syed, D. Kudithipudi, "Analysis of Wide and Deep Echo State Networks for Multiscale Spatiotemporal Time Series Forecasting," Proceedings of the ACM Neuro Inspired Computational Elements (NICE) Workshop, Albany, NY, 2019.

Z. Carmichael, H. Syed, S. Burtner, D. Kudithipudi, "Mod-DeepESN: Modular Deep Echo State Network," Annual Conference on Cognitive Computational Neuroscience, Philadelphia, PA, 2018.
arXiv

Z. Carmichael. "Towards Lightweight AI: Leveraging Stochasticity, Quantization, and Tensorization for Forecasting." Master's Thesis, Department of Computer Engineering, Rochester Institute of Technology. 2019.
RIT Scholar Works


Proceedings Paper | § Oral Presentation | Poster Presentation

Research Experience


Computer Vision Research Lab (CVRL) Graduate Research Assistant — Notre Dame, Notre Dame, IN

August 2019 - May 2024
  • Proposed a joint predictive and generative model for intrinsically interpretable classification. By combining concepts in explainable AI (prototypical neural networks) and generative AI (normalizing flows), we achieve state-of-the-art accuracy, density estimation, calibration quality, and interpretability across computer vision benchmarks
  • Conducted research demonstrating the infidelity of post hoc explanation methods for black box interpretation
  • Developed an open-source symbolic framework that enables researchers to study feature attribution, interaction effects, & explanations of arbitrarily complex models
  • Proposed a highly effective defense for explainers against adversarial attacks to identify malicious auditees and recover faithful explanations
  • Defended by Ph.D. dissertation in March 2024

Research Associate Intern — Hewlett Packard Enterprise (HPE) Labs, Milpitas, CA

May 2023 - March 2024
  • Developed methods for the evaluation and enhancement of natural and adversarial robustness in neural networks
  • Developed a neural surrogate for a computational fluid dynamics solver to improve data center sustainability, achieving 2,000× speedup. The surrogate is combined with online reinforcement learning for the optimization of carbon footprint in data centers

Research Intern — Mitsubishi Electric Research Laboratories, Cambridge, MA

June 2022 - September 2022
  • Conducted original research on intrinsically human-interpretable AI for vision tasks (prototypical part neural networks) under supervision of Dr. Mike Jones
  • Uncovered and mitigated a fundamental shortcoming of prototypical part neural networks that can produce highly misleading explanations - my solution (which led to a patent application that I helped author) quantitatively improves interpretability for this model which is often applied in high-stakes domains including biomedical imaging

Nu.AI Laboratory Research Fellow — UTSA, San Antonio, TX

August 2019 - May 2021

Nu.AI Laboratory Graduate Research Assistant — RIT, Rochester, NY

January 2018 - August 2019
  • Collaborated with epidemiologists & demographers in the modeling of COVID-19 infectious spread. Developed a live online dashboard for Texas state showcasing case data & forecasts
  • Researched the accuracy-energy-latency trade-off of network compression via low-precision arithmetic & custom hardware architecture
  • Improved efficiency of neural networks for time series forecasting upwards of 95% in size & training speed using randomness & compression for resource-constrained devices

Fellowships & Awards


NSF Graduate Fellowships Research Program (GRFP) Honorable Mention
2020
RIT Outstanding M.S. Thesis Award
2020

Thesis: "Towards Lightweight AI: Leveraging Stochasticity, Quantization, and Tensorization for Forecasting"

UTSA Best Poster: Fundamental Research in AI (Ph.D.)
2019

Poster: "Cheetah: Mixed Low-Precision Hardware & Software Co-Design Framework for DNNs on the Edge"

University of Notre Dame Jack and Mary Ann Remick Fellowship in Engineering
2019 - Present
University of Notre Dame Kilgallon Family Graduate Fellowship
2019 - Present
RIT KGCOE Dean's List
2014 - 2019
RIT Presidential Scholarship
2014 - 2018
RIT BS/MS Tuition Award
2014 - 2018
RIT Excellence in Computing
2014

Service


Computer Vision Foundation (CVF)
IEEE Access
IEEE Transactions on Computers
Czech Science Foundation
IEEE Transactions on Neural Networks and Learning Systems
Neural Information Processing Systems (NeurIPS) Workshop on Compact Deep Neural Networks

Membership


Institute of Electrical and Electronics Engineers (IEEE) Student Member
Oct 2018 - Present
Tau Beta Pi - The Engineering Honor Society
Oct 2018 - Present
The National Society of Leadership and Success
Oct 2018 - Present
Sigma Xi Nomination
Jun 2020

Skills


Languages

  • Python
  • C++
  • C
  • Java
  • LaTeX
  • HTML/CSS
  • Shell/Bash
  • JavaScript
  • TCL
  • System Verilog/Verilog
  • VHDL
  • ARM Assembly
  • Rust

Software and Libraries

  • TensorFlow
  • Scikit-Learn
  • PyTorch
  • Git
  • NumPy/SciPy
  • SVN
  • Docker
  • Singularity
  • Matplotlib/Seaborn
  • MATLAB/Octave
  • Django
  • Apache HTTPD
  • SQL
  • Xilinx ISE/Vivado
  • ROS
  • ModelSim
  • Keil μVision
  • Microsoft Office
  • Eclipse
  • Atlassian Suite

Hardware

  • Cortex M0+/M4 Architectures
  • NXP/Freescale KL46Z/KL64F
  • Arduino
  • Spartan6 FPGA

Operating Systems

  • Arch Linux
  • Debian/Ubuntu & variants
  • CentOS/RedHat
  • NixOS
  • Amazon Linux AMI
  • Windows

Other

  • System administration
  • Project management
  • Web development

Other Experience


Web Manager — Computer Vision Foundation (CVF)

September 2019 - Present
  • Position funded my PhD
  • Rewrote, audited, & actively maintain CVF Open Access to better serve papers, talks, posters, & other open content from the CVPR, ICCV, ECCV, WACV, & ACCV conferences to 500,000+ monthly visitors
  • Automated synchronization of CVF COVE computer vision datasets & arXiv erratum retrieval with Open Access
  • Discovered & mitigated several SQL security vulnerabilities
Languages
  • Python
  • SQL
  • Bash
Software
  • Apache HTTPD
  • Pandas
  • Web scraping tools

Graduate Teaching Assistant — University of Notre Dame; Rochester Institute of Technology

2018 - 2020
  • Notre Dame CSE-60625/40625 Advanced Topics in Machine Learning Teaching Assistant (>20 students)
  • Notre Dame CSE-30151 Theory Of Computing Teaching Assistant (>30 students)
  • RIT CMPE-380 Applied Programming Grader (>30 students)
  • RIT CMPE-250 Assembly Language Grader (>30 students)

Digital Engineering Intern — Plexus Corp., Raleigh, NC

June 2017 - August 2017
  • Carried out RTL design of FPGA-agnostic module for evaluation of FPGA cooling systems, validated all test cases with digital engineering team
  • Developed embedded software for a battery testing unit using the FRDM-K64F dev board, validated design & integration with mechanical, electrical, & software teams
Languages
  • C/C++
  • System Verilog
  • Powershell
Hardware
  • NXP/Freedom K64F
  • Xilinx Spartan FPGA

Software Engineering/Research Intern — CUBRC, Inc., Cheektowaga, NY

June 2016 - December 2016
  • Developed a machine learning framework to model surgery risk, patient mortality, & other analytics using TensorFlow & scikit-learn with automatic model search & hyperparameter optimization
  • Worked with customers in the design of electronic health record-unifying database & interface
Languages
  • Python {2, 3}
  • Shell/Bash
  • SPARQL and RDF
Software and Libraries
  • TensorFlow
  • Scikit-Learn
  • Flask
  • NumPy
  • Kafka
  • Atlassian Suite {JIRA, Confluence, HipChat, BitBucket}

Senior SD Representative and RA Publisher — ITS Service Desk (RIT), Rochester, NY

Fall 2014 - Spring 2016
  • Trained new representatives in customer interaction, internal procedures, and general troubleshooting methods
  • Worked with RIT staff, faculty, and students to resolve hardware and software problems, manage RIT computer accounts, and provide information on technology services
Software
  • Footprints
  • Microsoft Office
  • RightAnswers
  • Atlassian Confluence

Founding Partner — CPUtiful Tech, Killingworth, CT

Summer 2015 - Spring 2016

CPUtiful Tech was a company comprising four partners with the mutual interest of creating a consumer electronics retail company with a business platform based on personable interactions. The company was dissolved in 2016.

Activities


RIT Tennis Club Treasurer
2014 - 2017
RIT Intramural Co-Ed Volleyball Captain
2014 - 2017
RIT Intramural Co-Ed Flag Football Captain
2015 - 2016
RIT KanJam Executive Board Member
2014 - 2016
Drumming (Kit)
Tennis