Prof. Ran He
Speech Title: "Variational Image Analysis under Limited Computational Resource"
Image data tend to be high-dimensional and large-scale. When given infinite computational resource, machine learning algorithms can generate exact results (prohibitively expensive). Variational approximation methods arise from the use of a finite amount of processor time. These methods are often built on top of standard function approximators. In this talk, we introduce a group of variational inference and learning algorithms that scale to high-dimensional and large-scale image datasets. First, we address the linear approximation to learn robust and compact local features of image data, named ordinal measures. Second, we address the quadratic approximation of a family of loss functions that widely used in image analysis. Accordingly, a half-quadratic optimization framework is proposed for modeling sparsity, low-rank recovery and noise. Third, we introduce an Introspective Variational Autoencoders to approximate the posterior distribution, then we can generate high-resolution images from the learnt distribution, paving a way for analysis via synthesis.
Assoc. Prof. Mohd Nazri Ismail
Speech Title: "Development of Wireless Sensor Network (WSN) and Mobile Ad-Hoc Network (MANET) Communication for Military Operation and SAR (Search and Rescue) Operation"
The study investigates and develops components for implementing an effective military and SAR (Search & Rescue) acknowledge/information/communication in closed network architecture. Since military and SAR personnel are always on the move, the dissemination of knowledge/information/communication needs a mobile platform to accommodate mobility of people. The mobile and wireless network platform should be able to sustain the remoteness and seclusion of military operation areas. Communication is one of key problems of a military operation especially due to environmental constraints. This study proposes on establishing a future soldier and SAR communication device with mobile Wireless Sensor Network (WSN) and Mobile Ah-Hoc Network (MANET) to suit the infantry operations in the urban and rural areas. The operational areas are considered to restricted and challenging locations. Wireless sensor network (WSN) and Mobile Ah-Hoc Network (MANET) will become inexpensive and common over the next decade Thus, a thorough study is vital to develop the most suitable smart equipment and network requirements for Malaysia’s military and SAR eco-system. Finally, this study has successfully developed new low cost device prototype using WSN and MANET approach for Military and SAR operation. This approach is able to transmit death and location status, movement location status, health monitoring status to the base station.