Research

Publications

Selected publications by venue and year. Also on Google Scholar and ORCID.

Citations
730
h-index
15
i10-index
19

Google Scholar · updated

2026 3 papers

Fast 3D Surrogate Modeling for Data Center Thermal Management

S. Sarkar, A. Guillen-Perez, Z. J. Carmichael, A. Naug, R. M. Cam, V. Gundecha

AAAI Conference on Artificial Intelligence, 2026

[venue] cfdsurrogate-modelssustainability

Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?

P. Chlenski, Z. J. Carmichael, A. Warikoo, J. Shao, Y. Ye, O. Yang, V. Miglani, N. Bandi

ICML Mechanistic Interpretability Workshop, 2026 · Spotlight

[venue] llminterpretabilitysurrogate-models

A Preliminary Study on Explaining Risk of Code Changes Using LLM-based Prediction Models

Y. Liu, K. Jabre, R. Abreu, Z. J. Carmichael, V. Murali, A. Patel, J. Ge, W. Sun

AIware, 2026 · accepted

[venue] llmcode-reviewinterpretability
2025 1 paper

A Flexible Framework for Hyperparameter Optimization Using Homotopy and Surrogate Models

S. J. Abraham, K. D. G. Maduranga, J. Kinnison, Z. J. Carmichael, J. D. Hauenstein, W. J. Scheirer

Scientific Reports, 2025

[venue] hyperparameter-optimization
2024 5 papers

Explainable AI for High-Stakes Decision-Making

Z. J. Carmichael

Ph.D. Dissertation, University of Notre Dame, 2024

[PDF] [venue] xaidissertation

This Probably Looks Exactly Like That: An Invertible Prototypical Network

Z. J. Carmichael, T. Redgrave, D. G. Cedre, W. J. Scheirer

European Conference on Computer Vision (ECCV), 2024

[arXiv] [code] interpretabilityprototypical-networksnormalizing-flows

CFD Surrogates for Data Center Sustainability Using 3D U-Net Convolutional Neural Network

S. Sarkar, A. Guillen-Perez, Z. J. Carmichael, V. Gundecha, A. Naug

IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm), 2024

[venue] cfdsurrogate-modelssustainability

Benchmark Generation Framework with Custom Distortions for Image Classifier Robustness

S. Sarkar, A. Ramesh Babu, S. Mousavi, Z. J. Carmichael, V. Gundecha, S. Ghorbanpour, R. Luna Gutierrez, A. Guillen, A. Naug

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024

[venue] robustnessbenchmarks

Pixel-Grounded Prototypical Part Networks

Z. J. Carmichael, S. Lohit, A. Cherian, M. J. Jones, W. J. Scheirer

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024

[venue] interpretabilityprototypical-networks
2023 6 papers

Learning Debuggable Models Through Multi-Objective Neural Architecture Search

Z. J. Carmichael, T. Moon, S. A. Jacobs

International Conference on Automated Machine Learning (AutoML) Workshop, 2023

[arXiv] interpretabilityneural-architecture-search

HomOpt: A Homotopy-Based Hyperparameter Optimization Method

S. J. Abraham, K. D. G. Maduranga, J. Kinnison, Z. J. Carmichael, J. D. Hauenstein, W. J. Scheirer

Preprint, 2023 · extended journal version: Scientific Reports 2025

[arXiv] hyperparameter-optimization

Enhancing Data Center Sustainability with a 3D CNN-based CFD Surrogate Model

S. Sarkar, A. Guillen, Z. J. Carmichael, V. Gundecha, A. Naug, A. R. Babu

NeurIPS Workshop on Tackling Climate Change with Machine Learning, 2023

[PDF] cfdsurrogate-modelssustainability

How Well Do Feature-Additive Explainers Explain Feature-Additive Predictors?

Z. J. Carmichael, W. J. Scheirer

NeurIPS Workshop XAI in Action: Past, Present, and Future Applications, 2023

[arXiv] [code] xaievaluation

Unfooling Perturbation-Based Post Hoc Explainers

Z. J. Carmichael, W. J. Scheirer

AAAI Conference on Artificial Intelligence, 2023

[arXiv] [venue] [code] xaiadversarial-robustnessauditing

Motif Mining: Finding and Summarizing Remixed Image Content

W. Theisen, D. G. Cedre, Z. J. Carmichael, D. Moreira, T. Weninger, W. J. Scheirer

IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023

[arXiv] [venue] computer-visionsocial-media
2022 1 paper

TERSE: Tiny Encryptions and Really Speedy Execution for Post-Quantum Private Stream Aggregation

J. Takeshita, Z. J. Carmichael, R. Karl, T. Jung

EAI International Conference on Security and Privacy in Communication Networks (SecureComm), 2022

[PDF] cryptographypost-quantum
2021 3 papers

A Framework for Evaluating Post Hoc Feature-Additive Explainers

Z. J. Carmichael, W. J. Scheirer

arXiv preprint, 2021 · precursor to NeurIPS XAI Workshop 2023

[arXiv] xaievaluation

Adaptive Autonomy in Human-on-the-Loop Vision-Based Robotics Systems

S. Abraham, Z. J. Carmichael, S. Banerjee, R. VidalMata, A. Agrawal

Workshop on AI Engineering – Software Engineering for AI (WAIN), 2021

[arXiv] roboticshuman-on-the-loopvision

ALPS: Adaptive Quantization of Deep Neural Networks With GeneraLized PositS

H. F. Langroudi, V. Karia, Z. J. Carmichael, A. Zyarah, T. Pandit, J. L. Gustafson

IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshop, 2021

[venue] low-precisionquantization
2020 1 paper

SIRNet: Understanding Social Distancing Measures with Hybrid Neural Network Model for COVID-19 Infectious Spread

N. Soures, D. Chambers, Z. J. Carmichael, A. Daram, D. P. Shah, K. Clark, L. Potter, D. Kudithipudi

IJCAI Disease Computational Modeling Workshop, 2020

[arXiv] [project] epidemiologytime-series
2019 8 papers

Performance-Efficiency Trade-off of Low-Precision Numerical Formats in Deep Neural Networks

Z. J. Carmichael, H. F. Langroudi, C. Khazanov, J. Lillie, J. L. Gustafson, D. Kudithipudi

ACM Conference for Next Generation Arithmetic (CoNGA), 2019

[arXiv] [venue] low-precisionhardware

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

H. F. Langroudi, Z. J. Carmichael, D. Pastuch, D. Kudithipudi

arXiv preprint, 2019

[arXiv] low-precisionhardware

Stochastic Tucker-Decomposed Recurrent Neural Networks for Forecasting

Z. J. Carmichael, D. Kudithipudi

IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2019

[PDF] tensor-decompositionforecasting

Deep Positron: A Deep Neural Network Using Posit Number System

Z. J. Carmichael, H. F. Langroudi, C. Khazanov, J. Lillie, J. L. Gustafson, D. Kudithipudi

IEEE Conference and Exhibition on Design, Automation and Test in Europe (DATE), 2019 · 24% acceptance rate

[arXiv] low-precisionhardware

Deep Learning Training on the Edge with Low-Precision Posits

H. F. Langroudi, Z. J. Carmichael, D. Kudithipudi

arXiv preprint, 2019

[arXiv] low-precisionhardware

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

Z. J. Carmichael

M.S. Thesis, Rochester Institute of Technology, 2019 · Awarded RIT Outstanding M.S. Thesis (2020)

[PDF] thesis

PositNN Framework: Tapered Precision Deep Learning Inference for the Edge

H. F. Langroudi, Z. J. Carmichael, J. L. Gustafson, D. Kudithipudi

IEEE Space Computing Conference (SCC), 2019

[venue] low-precisionhardware

Analysis of Wide and Deep Echo State Networks for Multiscale Spatiotemporal Time Series Forecasting

Z. J. Carmichael, H. Syed, D. Kudithipudi

ACM Neuro Inspired Computational Elements (NICE) Workshop, 2019

[venue] reservoir-computingforecasting
2018 1 paper

Mod-DeepESN: Modular Deep Echo State Network

Z. J. Carmichael, H. Syed, S. Burtner, D. Kudithipudi

Annual Conference on Cognitive Computational Neuroscience, 2018

[arXiv] reservoir-computing