Sensors In-home for Elder Wellbeing 

In collaboration with Sengkang General Hospital and National University of Singapore, Nov 2019 - Nov 2024.

This project, led by A/Prof Iris Rawtaer (SKH) aims to utilise multimodal sensor networks for early detection of cognitive decline. Under this project, the SKH and NUS team will oversee the project operations, screening recruitment, psychometric evaluation, data analysis, data interpretation, reporting and answer of clinical research hypotheses. The SMU team will collaborate with SKH and NUS to provide technical expertise for this study by ensuring safe implementation and maintenance of the sensors in the homes of the participants, provide the sensor obtained data to the clinical team and apply artificial intelligence methods for predictive modelling.

Trustworthy Federated Ubiquitous Learning

Funded by the National Research Foundation, Singapore under its AI Singapore Programme (AISG Award No: AISG2-RP-2020-019), Apr 2021 - Mar 2025.

Federated Learning (FL), in which training happens where data are stored and only model parameters leave the data silos, can help AI thrive in the privacy-focused regulatory environment. FL involves self-interested data owners collaboratively training machine learning models. In this way, end-users can become co-creators of AI solutions. To enable open collaboration among FL co-creators and enhance adoption of the federated learning paradigm, this project aims to develop the Trustworthy Federated Ubiquitous Learning (TrustFUL) framework, which will enable communities of data owners to self-organize during FL model training based on three notions of trust: 1) trust through transparency, 2) trust through fairness and 3) trust through robustness, without exposing sensitive local data. As a technology showcase, we will translate TrustFUL into an FL-powered AI model crowdsourcing platform to support AI solution co-creation.

Enhancing Generalizability and Explainability of Multi-Agent Reinforcement Learning

In collaboration with DSO National Laboratories Oct 2024 - Sep 2026. 

This project shall (i) enhance the generalizability of hierarchical multi-agent learning and control framework for heterogeneous agents in a range of scenarios and (ii) develop algorithms to analyse and explain the learned behaviour models at the various levels.

Knowledge Extraction, Modelling and Explainable Reasoning for General Expertise- Work Package 2. 

Funded by RIE2020 Advanced Manufacturing and Engineering (AME) Programmatic Grant, Feb 2020 - Jan 2023. 

As part of the K-EMERGE (Knowledge Extraction, Modelling, and Explainable Reasoning for General Expertise) project funded by A*STAR under the RIE2020 Advanced Manufacturing and Engineering (AME) Programmatic Grant, this project aims to develop computational models and technologies for representing and learning domain knowledge extracted from text-based technical documents.

Funded by Singapore Ministry of Education Academic Research Fund (AcRF) Tier-1, Apr 2020 - Mar 2022.

In this project, we shall investigate a new breed of advanced information systems, known as Cognitive Information Systems, which incorporate human-like cognitive capabilities for information processing.  Specifically, the project shall review the state-of-the art technologies in artificial intelligence, cognitive science and neural computing, and study the research challenges faced in building cognitive information systems. A main goal of the project is to develop a framework outlining the general structure and characteristics of such information systems. To this end, this project shall explore how a family of biologically-inspired self-organizing neural networks known as fusion Adaptive Resonance Theory (fusion ART), encompassing a set of universal neural coding and adaptation principles, may be used as a building block of cognitive information systems.