UDMuhammad Umair DanishApplied AI · Western University
Research collaborations

Interdisciplinary by design.
Rigorous by practice.

I lead applied AI research by pairing methodological depth with the domain expertise required to make models useful, interpretable, and trustworthy.

Research leadership Core collaborators
Western UniversityWilfrid Laurier UniversityZayed UniversityMcGill University
Muhammad Umair Danish in Western University doctoral regalia
Research role Principal Investigator
& Research Lead
Research interfaces AI · HCI · Energy · Health · Music · Security
Portrait of Muhammad Umair Danish
Principal InvestigatorPersonal applied-AI research programme
Research lead

Muhammad Umair Danish, PhD

Postdoctoral Fellow in Applied AI · Western University

I shape the research agenda, build the methodological architecture, lead model development and validation, and coordinate the translation of technical findings into human and domain outcomes.

01
Scientific direction

Mechanistic interpretability, human-centered XAI, reliable temporal learning, and generative-AI evaluation.

02
Methodological rigor

Controlled baselines, ablations, statistical validation, reproducible implementation, and failure analysis.

03
Interdisciplinary translation

Connecting AI methods to energy, finance, imaging, music therapy, health, and cybersecurity.

04
Research communication

Publication strategy, visual explanation, open software, and collaborative scholarly development.

Explore research View publications Discuss collaboration
Core collaborators

Complementary expertise.
Shared standards of evidence.

These collaborations are central to my current research programme and connect technical AI development with supervision, human-centered evaluation, clinical interpretation, and domain-grounded study design.

Katarina Grolinger01
PhD & Postdoctoral SupervisorWestern University

Katarina Grolinger, PhD, P.Eng.

Canada Research Chair in Engineering Applications of Machine Learning · Associate Professor

Dr. Grolinger provides long-term scientific supervision across my doctoral and postdoctoral research, with a shared focus on reliable machine learning, energy applications, interpretability, and rigorous empirical evaluation.

Research relationship

Research supervision, methodological direction, energy forecasting, physics-guided learning, interpretability, and publication development.

Research supervisionMachine learningEnergy AIInterpretability
Official university profile
Umair Rehman02
Major Research CollaboratorWestern University

Umair Rehman, PhD

Assistant Professor · Director, Human-Centered Computing Group

Our collaboration combines explainable AI with human-computer interaction and cognitive engineering to study not only how models explain decisions, but how people understand, trust, and use those explanations.

Selected shared work

Linear Lens, GLIPS, unified evaluation of AI-generated images, and ongoing human-centered XAI research.

Human-centered AIHCIMechanistic interpretabilityPerceptual evaluation
Official university profile
Demian Kogutek03
Computational Music Therapy CollaboratorWilfrid Laurier University

Demian Kogutek, PhD, MTA

Assistant Professor, Music Therapy · Director, Conrad Institute for Music Therapy Research

Our interdisciplinary work translates improvisational music-therapy practice into reproducible computational analysis while preserving the clinical meaning of timing, interaction, and musical response.

Selected shared work

MidiPy and advanced correlation analysis for improvised active music-therapy sessions and scripts.

MidiPyMusic therapyMIDI analyticsClinical research tools
Official university profile
Aleksandra Zecevic04
Human Factors & Health CollaboratorWestern University

Aleksandra Zecevic, PhD

Professor · School of Health Studies

This collaboration connects human factors, safety, aging, and health research with the evaluation of human-facing AI explanations and the decisions people make when assisted by intelligent systems.

Current collaboration

Human–AI interaction studies examining how textual, graphical, and interactive explanations affect trust, understanding, cognitive response, and decision making.

Human factorsHuman–AI interactionTrustDecision making
Official university profile
Collaboration portfolio

Shared work, clearly connected.

Select a programme to see how complementary expertise contributes to the research.

01
Mechanistic interpretability

Linear Lens

A non-interventional, human-centered approach for explaining internal neural representations without modifying deployed models.

Muhammad Umair DanishUmair RehmanKatarina Grolinger
View publication page
Extended research network

Security, privacy, and high-stakes analytics.

Senior collaborators extend the programme into cybersecurity, digital forensics, privacy-aware analytics, and complex evidence environments.

Farkhund Iqbal
AI & Cybersecurity Collaborator

Farkhund Iqbal, PhD

Professor, Zayed University. Shared interests include explainable machine learning, digital forensics, cybersecurity, and financial-credit evaluation.

Official profile ↗
Benjamin Fung
Data Mining & Security Collaborator

Benjamin C. M. Fung, PhD, P.Eng.

Professor, McGill University. The research connection spans data mining, privacy protection, cybersecurity, and evidence-driven AI systems.

Official profile ↗
Collaboration principles

Strong collaborations make the evidence stronger.

01Define the real question

Technical objectives are aligned with the human, scientific, or domain decision the work must support.

02Design evidence early

Baselines, outcomes, validation criteria, and interpretation plans are established before conclusions are drawn.

03Build reproducibly

Code, assumptions, datasets, and failure modes are made explicit so results can be examined and extended.

New collaborations

Building interpretable AI for real decisions?

I welcome research conversations where methodological rigor and domain relevance are equally important.

Start a research conversation