Research Scientist (1-year contract), SCIS

Job Description

The candidate will be responsible for leading and conducting research on (a) developing new SNN architectures and training techniques to support multi-modal inference, especially when the sensor streams have different event sampling rates and kinematic sensitivity, and (b) applying the collaborative sensing paradigm (pioneered by PI Misra's group) to support energy-efficient correlation of features and insights from sensors on multiple on-body wearable devices. Successful candidate will be part of an active research team led by Prof Archan Misra from School of Computing and Information Systems, Singapore Management University.

• Developing multi-modal feature alignment and stateful cross-modal fusion based techniques for SNN-based inference.
• Adapting the core principles of collaborative sensing to devise SNN-based algorithms for wireless based transfer of spatiotemporal summaries of hidden SNN states across multiple SNN instances.
• Work with other team members to collect suitable wearable event sensor data and devise preprocessing techniques to ensure model robustness.
• Provide supervision to a small team of researchers to enable the submission of 1-2 high-quality research publications annually.
• Other duties as assigned.

Qualifications

• Minimum Ph.D. in Computer Science/Electrical Engineering or relevant disciplines from a reputable institute, with particular focus on optimizing AI execution on IoT or edge devices.
• 3+ years project experience with multi-modal spatial analysis, involving RGB and LIDAR-generated data.
• Coding experience with DNN and vision AI frameworks, such as pyTorch, TensorFlow and/or OpenCV, is essential.
• Professional experience in edge AI optimization and collaborative execution of DNN-based inference across multiple low-resource edge platforms is essential.
• Project-level experience with DNN-based models (CNNs, transfers) for object detection using both RGB images and LIDAR 3D point cloud data is strongly desired.
• Ability to work cooperatively as part of a small, agile academic research team is essential.
• High-quality publications in high-quality academic conferences and journals, especially in embedded sensing and ubiquitous computing venues is highly preferred.

The candidate will be responsible for leading and conducting research on (a) developing new SNN architectures and training techniques to support multi-modal inference, especially when the sensor streams have different event sampling rates and kinematic sensitivity, and (b) applying the collaborative sensing paradigm (pioneered by PI Misra's group) to support energy-efficient correlation of features and insights from sensors on multiple on-body wearable devices. Successful candidate will be part of an active research team led by Prof Archan Misra from School of Computing and Information Systems, Singapore Management University.

• Developing multi-modal feature alignment and stateful cross-modal fusion based techniques for SNN-based inference.
• Adapting the core principles of collaborative sensing to devise SNN-based algorithms for wireless based transfer of spatiotemporal summaries of hidden SNN states across multiple SNN instances.
• Work with other team members to collect suitable wearable event sensor data and devise preprocessing techniques to ensure model robustness.
• Provide supervision to a small team of researchers to enable the submission of 1-2 high-quality research publications annually.
• Other duties as assigned.