Probabilistic graphical models and Causal AI

The project intends to develop methods to suggest how to optimally intervene so that the future state of the system will best suit our interests. The power of probabilistic graphical models to model complex relationships and interactions among a large number of variables facilitates many applications. However, such models only aim to understand the underlying environment. What is ultimately needed in many real-world applications is to suggest how we ought to intervene or act, so as to alter the environment to best suit our interests. The proposed project aims to achieve this using probabilistic graphical models on massive real-world data sets, thus facilitating a variety of applications from health care to commerce and the environment.

A/Prof Qinfeng Shi; Assistant Professor Julian McAuley; Associate Professor Pawan Mudigonda

Probabilistic Graphical Models (PGMs) use graphs to represent the interactions between random variables and provide a formalism by which to represent complex probabilistic relationships. Despite the success of PGMs in many fields, the learning on real industrial large scale applications is very slow. I will exploit the sparsity and compressibility in PGMs, and turn the large scale PGMs to a number of small scale PGMs. Solving these small scale PGMs and then reversely recover the solutions in the original large scale PGMs in the context of Compressive Sensing. This way, I can effectively deal with large scale PGMs in the computational complexity of small scale PGMs as well as provide theoretical guarantees on the consistency of the solution.

A/Prof Qinfeng Shi

• Accelerate the discovery of clean energy materials and sustainable technologies.
• Optimise industrial operations by identifying true efficiency drivers.
• Improve manufacturing resilience by modelling intervention impacts on safety, quality, and productivity.
• Support health and sport science through causal insights into training, performance, and wellbeing.

• Uncover the root causes of bottlenecks and disruptions in global supply chains.
• Model the cascading impacts of shocks (e.g., pandemics, conflicts, extreme weather) on trade and logistics.
• Test intervention strategies such as diversification of suppliers or rerouting of logistics.
• Support resilience planning for critical infrastructure such as energy grids, transport networks, and digital systems.

• Distinguish causal factors in disease progression, treatment effectiveness, and patient outcomes.
• Personalise treatment strategies by predicting the impact of medical interventions.
• Improve public health planning by modelling the effects of policies on populations.
• Build robust, interpretable models that doctors and policymakers can trust.

• Detect crop stress and soil changes before visible damage occurs.
• Optimise fertiliser, irrigation, and planting strategies for sustainable yield.
• Model climate-driven risks such as drought, pests, and disease outbreaks.
• Support food security through resilient farming systems.

• Reveal hidden drivers of harmful algal blooms and water contamination.
• Model ecological responses to policy changes or land use interventions.
• Provide early warnings for floods, fires, and other natural hazards.
• Support integrated catchment management and resilient water systems.

• Identify leverage points for emissions reduction and biodiversity protection.
• Explore sustainable development pathways under different policy or investment scenarios.
• Model land restoration strategies to optimise ecological and economic outcomes.
• Integrate satellite, climate, and policy data for systems-level planning.

• Anticipate systemic risks across finance, environment, and geopolitics.
• Provide robust evidence for policy and governance under uncertainty.
• Design interventions that reduce vulnerabilities in critical national systems.
• Support ethical and responsible AI deployment by exposing hidden biases and unintended consequences.