The Distributed Learning and Orchestration (DENOS) Lab at the University of Calgary was founded in 2021 by Dr. Steve Drew, Assistant Professor in the Department of Electrical and Software Engineering. We work on distributed learning, agentic simulation and reasoning, building intelligent, sustainable, and privacy-preserving systems that span cloud-edge infrastructures.
Much of our work puts large language model agents to work on real problems in health and public safety. We build multi-agent systems that simulate emergency department operations, grade clinical summaries, share health data under verifiable consent, and model wildfire evacuations. In parallel, we design federated learning methods that keep cloud-edge services resilient and carbon-aware while protecting sensitive data.
We focus on moving from theory to deployment, with methods that stay trustworthy and reproducible, minimize energy and carbon footprint, and hold up at scale. Our work has earned a Best Paper Award and national research funding and appears in venues across machine learning, systems, and digital medicine.
DENOS Lab is recruiting motivated MSc, PhD, and post-doctoral researchers in machine learning, distributed systems, and AI + Health. Prospective students should apply to a graduate program in the Department of Electrical and Software Engineering and mention Dr. Drew in their application.
DENOS Lab pursues resilient, sustainable, and privacy-preserving systems for distributed learning, agentic simulation and reasoning. Six flagship directions anchor our current work.
ED Sim is a large language model simulation of an emergency department, with autonomous agents for patients, nurses, and physicians. Built with Dr. Jessalyn Holodinsky, it lets teams find bottlenecks and test changes without disrupting real operations, aiming to cut wait times and clinician burnout.
STARFISH is a secure, interoperable framework for global health data sharing that lets individuals control their data while enabling compliant research. Its agentic AI protocol exchanges data between personal and institutional agents using verifiable consent, cryptographically signed mandates, the Agent Payments Protocol (AP2), and blockchain auditability, with analytics run inside a secure computation container.
AgentEvac is an agentic simulator for wildfire evacuations that couples SUMO traffic simulation with LLM-driven agents following the Protective Action Decision Model (PADM). By encoding psychologically grounded, adaptive behavior, it lets us test warning strategies and route policies and measure how uncertainty and social trust shape safety and efficiency.
SENSE supports the emotional wellness of nurses and social workers, pairing wearable sensor data with machine learning to help frontline staff manage stress in real time. Its Apple Watch stress monitoring assistant streams heart rate through HealthKit, computes HRV features from beat-to-beat intervals, and runs a Siamese LSTM model to flag elevated stress as it rises.
LENS is a role-aware multi-agent pipeline that grades clinical summaries. Three agents standing in for a physician, a triage nurse, and a bedside nurse score each summary across eight rubric dimensions, and an Orchestrator Disagreement view shows where their judgments diverge.
As federated learning (FL) spreads across heterogeneous infrastructures, slow convergence drives excessive energy use on cloud and battery-powered edge devices. We design FL methods that budget clients by availability and carbon footprint and guide cloud-edge orchestration toward resilient, sustainable, real-world deployment.
Dr. Steve Drew and the DENOS Lab are featured by the University of Calgary for ED Sim, an AI simulation platform whose autonomous agents model patients, staff, and workflows so that emergency departments can find bottlenecks and test solutions without disrupting real operations.
Read the feature →
DENOS Lab research received a Best Paper Award at the 2025 IEEE International Conference on Autonomous and Trusted Computing (ATC), held in Calgary as part of the IEEE Smart World Congress.
Visit ATC 2025 →
Dr. Steve Drew and co-principal investigator Dr. Jessalyn Holodinsky won a New Frontiers in Research Fund (NFRF) Exploration grant to optimize emergency department operations with multi-agent large language models, pairing rapid patient-history summarization with a simulation of clinical workflows so that hospitals can shorten wait times and ease clinician burnout.
Read the announcement →
A UCalgary and Michigan State University team including Dr. Steve Drew and Fan Dong won third place in the Privacy-Enhancing Technologies (PETs) Prize Challenge at the 2023 Summit for Democracy, using a federated learning approach that detects financial crime across banks without exposing personal data. Congratulations, Fan Dong!
Read the story →
Lab Director · Assistant Professor, Department of Electrical and Software Engineering, University of Calgary
Dr. Steve Drew leads the DENOS Lab. His research spans federated learning, edge computing, distributed systems, blockchain, and AI for health. He serves as an Associate Editor of the IEEE Internet of Things Journal and Neurocomputing.
Find Dr. Drew on Google Scholar and his UCalgary Profile.
Postdoctoral Researcher, University of Calgary
Dr. Vincent Michalski is a postdoctoral researcher in the DENOS Lab, working on representation learning and reasoning.
Find Vincent on Google Scholar.
PhD Student in Electrical and Software Engineering, University of Calgary
PhD Candidate in Electrical and Software Engineering, University of Calgary
PhD Student in Electrical and Software Engineering, University of Calgary
PhD Student in Electrical and Software Engineering, University of Calgary
PhD Student in Electrical and Software Engineering, University of Calgary
PhD Candidate in Electrical and Software Engineering, University of Calgary
PhD Student in Electrical and Software Engineering, University of Calgary
Master Student in Electrical and Software Engineering, University of Calgary
Master Student in Electrical and Software Engineering, University of Calgary
Master Student in Electrical and Software Engineering, University of Calgary
Master Student in Electrical and Software Engineering, University of Calgary
Undergraduate Intern, University of Calgary
Summer Intern at 2025.
Research Intern, University of Calgary
Summer Intern at 2025.
Master Student 2022 - 2024. currently Ph.D. student at University of South California.
Master Student 2022 - 2024.
Undergraduate Intern, University of Calgary
Master Student 2023 - 2025. First employment as Lead Software Engineer at Broadbill Energy Inc.
We collaborate across disciplines and institutions to move our research into practice.
Selected work from DENOS Lab on federated learning, edge computing, trustworthy AI, and health informatics. The full and up-to-date list is on Google Scholar.
View all publications on Google Scholar
COLA-GLM: Collaborative one-shot and lossless algorithms of generalized linear models for decentralized observational healthcare data
Topology-aware Federated Learning in Edge Computing: A Comprehensive Survey
FedGreen: Carbon-aware Federated Learning with Model Size Adaptation
FedLE: Federated Learning Client Selection with Lifespan Extension for Edge IoT Networks
USDNL: Uncertainty-based Single Dropout in Noisy Label Learning
Resilient and Communication Efficient Learning for Heterogeneous Federated Systems
Shoreline: Data-Driven Threshold Estimation of the Online Reserves of Cryptocurrency Trading Platforms
Distributed Data Vending on Blockchain
EdgeChain: Blockchain-based Multi-vendor Mobile Edge Application Placement
Statutory holidays and key dates for the 2026 academic year in Canada.
