Nayyar Zaidi

Research Focus

Generative Tabular Models

Synthetic data generation for structured tabular data, with emphasis on dependency modelling, fairness, privacy, data scarcity, and trustworthy generation.

Learn more »

Knowledge-guided Learning

Machine learning methods that integrate domain knowledge, knowledge graphs, constraints, scientific principles, and expert structure.

Learn more »

Multimodal Learning

Robust multimodal learning and fusion methods, particularly for medical image analysis and heterogeneous data sources.

Learn more »

Anomaly Detection

Structure-aware and dependency-level anomaly detection for structured and heterogeneous tabular data.

Learn more »

Tabular Classification

Classification methods for tabular data, including feature interactions, Bayesian network classifiers, causal learning, and few-pass learning.

Learn more »

Recent Updates

• October 2026: Three papers from our research group have been accepted at NeurIPS 2026 Workshops in Sydney, Australia! Congratulations to Anjum Asiya and Shashwat Suthar on their excellent work. The accepted papers are:
  • Correct Memory, Changed Assumptions: Procedural Transfer Across Trust Boundaries in Coding Agents — Third Workshop on Agents in the Wild: Safety, Security, and Beyond.
  • Same Outcome, Different Behavior: Interpreting Procedural Transfer Across Trust Boundaries in Coding Agents — First Workshop on Interpreting Agent Behavior (IAB).
  • Identifying Trajectory Specific Command Uptake in Software Engineering Agents: Evidence from Cue Exclusion Controls — Workshop on Who Verifies the Agents? Toward Reliable Agent Development.
• 2026: Two of our papers were accepted to the ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), both as oral presentations. The papers are Effective and Robust Multimodal Medical Image Analysis and Advancing Multimodal Fusion on Heterogeneous Medical Data with Hybrid Geometry Attention.
• 2026: Our work on adversarial robustness, Certified vs. Empirical Adversarial Robustness via Hybrid Convolutions with Attention Stochasticity, was accepted to the International Conference on Learning Representations (ICLR).
• 2026: Our paper Does a Hybrid Space-Aware Randomized Defense Improve Empirical and Certified Adversarial Robustness? was accepted to the International Conference on Machine Learning (ICML).
• 2026: Our paper HiGMR-Net: Hierarchical Graph Memory Refinement Network for Multi-View EEG Connectivity Sequence Learning was accepted to the ACM International Conference on Multimedia (ACM MM), to be held in Rio de Janeiro, Brazil.
• 2026: Our first paper in Medical Image Computing and Computer Assisted Intervention (MICCAI), Prompt and Probe: Data-Efficient Adaptation of Black-Box Foundation Models via Unified Active Visual Prompting, was accepted.
• 2026: My student Uzma Shafiq’s survey paper, A Survey on the Use of Machine Learning within Computer Algebra, was accepted in ACM Computing Surveys.
• 2025: Our group was awarded an Alfred Deakin Postdoctoral Research Fellowship, with Dr Christine Dewi joining the DMKD Group as an Alfred Deakin Postdoctoral Research Fellow. Christine’s work strengthens our research in knowledge-guided learning and computer vision.
• July 2024: Started serving as an Associate Editor for Big Data Research.
• November 2023: Delivered an invited talk, Can We Trust Machine Learning Models?, at the Deakin Cyber Conference, Melbourne. [PDF]
• October 2023: Delivered an invited talk, Can We Trust Machine Learning Models? Who Watches the Watchers?, at the Australian Cyber Conference, Melbourne. CyberCon 2023

Prospective PhD Students

Scholarships at Deakin are limited and highly competitive. Many applicants have strong academic records and publications in leading venues. If you are interested in working with me, I recommend contacting me well before your intended start date.

Prospective students are encouraged to develop a clear research proposal aligned with one of my research areas. Strong preparation may include prior research experience, a relevant publication record, and evidence of technical depth in computer science, mathematics, engineering, data science, or a closely related discipline.

If you are interested in joining my group, please contact me with your CV, academic transcripts, publication list, and a brief statement of research interests. Students who already hold an externally funded scholarship are also welcome to contact me directly.

Available Positions

Current advertised PhD scholarship and position opportunities are listed below. This section will be updated when new funded PhD projects, scholarships, visiting positions, or research roles become available.

• Deakin RTP and DUPRS Scholarships: Main Deakin research scholarship page

Invited Talks

• August 2023: Deep Generative Models are Great for Structured Data — Well, What About Tabular Data? Invited Talk, First International Conference on AI-generated Content (AIGC 2023) , Shanghai, China.
• 2023: Can We Trust Machine Learning Models? Invited Talk, Deakin Cyber Conference, Melbourne, Australia. [Slides]
• 2023: Can We Trust Machine Learning Models? Who Watches the Watchers? Invited Talk, Australian Cyber Conference (CyberCon 2023) , Melbourne, Australia. [Conference Programme]