As part of the celebration of the 16th anniversary of the founding of the University, the Faculty of Information Technology has organized three seminars composing of talks by distinguished researchers from academia and industry.
1st Seminar
Date : 8th March 2016
The program was as follows:
Speaker: Prof. Jiannong Cao, Hong Kong Polytechnic University
Topic: Smart Computing Drives Innovation
Abstract: Smart computing is emerging as an important multidisciplinary area, which effectively makes use of computer hardware, software, social media and communication networks together with digital sensors, smart devices, internet technologies, big data analytics, computational intelligence and intelligent systems to realize various innovative applications. It can be broadly classified into two major areas: how to design and build smart computing systems and how to use computing technology to design smart things and make human life better. The seminar will provide an overview of the research works in smart computing in the Department of Computing, The Hong Kong Polytechnic University, with a three-tier research framework. Tier 1 serves to focus our overall research efforts to two department-driven themes, namely "Big Data Computing" and "Human-Centred Computing". Tier 2, the backbone of Tier 1, represents our faculty members' six expertise areas. Tier 3 consists of cross-area research centres and laboratories established by our staff in collaboration with world-renowned researchers, facilitating research with cutting-edge projects and technologies.
Speaker: Prof. Yu Pei, Hong Kong Polytechnic University
Topic: Automatic Fixing of Programs with Contracts
Abstract: Debugging---the activity of finding and correcting errors in programs---is so every day in every programmer's job that any improvement at automating even parts of it has the potential for a significant impact on productivity and software quality. While automation remains formidably difficult in general, the last few years have seen the first successful attempts at automatically generating fixes to errors in some situations.
In this talk, I will present techniques and a supporting tool, collectively referred to as Auto Fix that programmers can use to automatically fix errors in object-oriented programs with contracts (a.k.a. assertions). Auto Fix takes a group of passing and failing test cases as input, with the failing tests revealing the fault to fix; it then analyzes the execution of the input tests, generates candidate fixes to the fault, validates the candidate fixes against a regression test suite, and ranks the valid fixes by preference before reporting them to the user. In the experiments conducted to evaluate Auto Fix, it generated fixes that are genuine corrections of quality comparable to those competent programmers would write to 25% of the subject faults. Auto Fix is integrated into the Eiffel Studio IDE and functions like a recommendation system that is capable of automatically finding bugs and suggesting fixes in the form of source-code patches.
Speaker: Prof. Xiaoxia Huang, Shenzhen Institutes of Advanced Technology
Topic: Visualizing and Leveraging Wireless Environment
Abstract: With the ubiquitous deployment of wireless communication technologies, we are now surrounded by wireless signals everywhere. Understanding and leveraging wireless environment are especially interesting issues which have received many attention recently. In this talk, the speaker introduces her work on urban spectrum measurement with our portable low cost device. Mounting the device on vehicles, fine-grained spectrum map of the city can be collected which reveals opportunities in improving spectrum efficiency. Besides helping improving communication efficiency, we also investigate how to harvest energy from wireless environment from the perspective of improving energy utilization.
2nd Seminar
Date : 15th March 2016
The program was as follows:
Speaker: Prof. Prof. Francis Chin, Hang Seng Management College
Topic: Multi-Task Machine Learning with Prior Knowledge on Related Data
Abstract: Machine learning is a subfield of Artificial Intelligence, which is in turn an area within Computer Science. Machine learning focuses on creating machines or systems that can learn from examples (training data) and make predictions on given data. Recently, there have been breakthroughs in machine learning due to the increase of computing power to allow the processing of large amounts of data (big data). Two fundamental supervised learning techniques: linear regression and the neural network will be discussed. Linear regression assumes that the outputs are linearly dependent on the input parameters (similar to assuming a set of points on a 2D plane lie on a straight line). The neural network is based on an old concept (perception) of simulating billions of interconnecting brain cells (neurons) to specify the mapping. Linear regression is simple and basic, while the neural network is promising especially when the network has many neurons and many layers of neurons (deep learning) and is equipped with nonlinear activation functions.
Our objective is to improve the performance of these two fundamental learning techniques with extra knowledge about the relationships between the inputs and outputs. In many applications, although the exact relationship between the inputs and outputs are unknown, the knowledge (or the assumption) that some of the inputs and outputs might be related may already be sufficient to improve the performance of these two learning techniques.
Speaker: Prof. Moshe Zukerman, City University of Hong Kong
Topic: Considering Cost and Network Survivability in laying telecommunications cables
Abstract: There have been costly consequences of several earthquakes and these events have raised concerns of possible Internet shutdown in certain areas. Of particular concern are situations where undersea cables lie close to each other in earthquake-prone areas. This presentation will report progress in a Collaborative Research Fund Project supported by the Research Grants Council of Hong Kong. The project aims to develop design methodologies for cabling systems taking into account cable attributes such as shape, shield, location, and capacity, trading off costs and network resilience. We integrate data on topography, earthquake effects and hazards into a network survivability model involving optimization over manifolds, leading to Pareto front results that trade off cost versus network failure risk. The optimization is based on evaluation of cost and network survival probability. Separation of cables in earthquake prone areas increases cost but also increases network survival probability. We aim to provide cable laying design solutions that minimize cost for a given network survival probability and also solutions that maximize network survival probability for a given budget. Results based on a combination of path planning for cable laying optimization taking account of the above mentioned earthquake effects and hazards using theoretical models as well as real world data will be presented.
Speaker: Dr. Yinian Zhu, Northwestern University
Topic: Applications of laser ultrasonic technique on nondestructive testing and evaluation of materials
Abstract: Intelligent Structural Health Management (ISHM) is the damage assessment for a process of continuous monitoring in engineering structural critical security, and it is to use diagnostic data for the estimating of structural reliability and potential longevity, and for the providing of appropriate remedies. As the part of the diagnostic systems, optical fiber sensors have played an important role in long-term structural health monitoring. The frequency caused by the energy that is released from structural crack growth and fracture in a material is near the ultrasonic wave frequency. To detect those waves, the fiber Bragg grating (FBG) as a sensing element has been considered as one of the most advanced and rapid testing technology so far. Because the resonance wavelength of the FBG can be shifted by dynamic strain that is created by ultrasonic wave, thus a dynamic strain demodulator is needed for frequency modulation of the FBG. This talk describes the development of three types of our home-made dynamic strain demodulators by using FBGs, which have been applied in the field to obtain the objective results.
Speaker: Prof. Yang Zhao, Laser Institute of Shandong Academy of Sciences, Shandong
Topic: Applications of Fiber Bragg Grating Sensing Technology in Structural Health Monitoring
Abstract: Nondestructive testing (NDT) is a wide group of analysis techniques used in science and industry to evaluate the properties of a material, component or system without causing damage. As one of unconventional NDT technology, laser ultrasonic technique (LUT) have played an important role in inspecting the materials in some extreme environments, such as high temperature, high pressure, corrosion, radioactive and etc. LUT uses lasers to generate and detect ultrasonic waves in a noncontact way. Thus, compared with the traditional ultrasonic technique, LUT has the advantages of high precision, high sensitivity without requirement for geometry of target material. So far, LUT has been well applied to composite inspections for the aerospace industry and on-line hot tube thickness measurements for the metallurgical industry. This talk describes the application of LUT on flaws detection in continuous casting slab and high-speed rail, and characterization of thermal barrier coating.
Speaker: Prof. Han Zhang, Shenzhen University
Topic: Nonlinear and ultrafast photonics based on two dimensional materials
Abstract: The graphene, topological insulator, molybdenum sulfide and black phosphorus represented as two-dimensional material were considered to be the hotspots of international academic circles. Two-dimensional material with unique physical properties, including a wealth of electronic energy band characteristics (from insulators, semiconductors to a metallic state, etc.), manually controllable electronic properties, quantum size effect, excellent mechanical properties, etc., indicate that two-dimensional materials have an extraordinary potential towards the construction of new optoelectronic devices.
The highlighted results obtained are given as follows:1) nonlinear optical properties of topological insulators and its applications: topological insulator is not only a broadband saturated absorbent material but also has a huge advantage Kerr effect, high damage threshold and so on. 2) In the two-dimensional nonlinear optical properties of MoS2 material and its applications to study the small layer MoS2 nanoplatelets from visible to near infrared absorption characteristics of nonlinear, it was found that a tiny thin layer of molybdenum disulfide together with its broadband saturated absorption effect makes it employable in its clamping applications in fiber lasers. 3) In the study of the optical and optoelectronic properties of the new two-dimensional material in terms of the black phosphorus, successfully established a new method for preparing the liquid material for the new two-dimensional black phosphorus, and further reveals the unique layers of black phosphorus according to Raman property and optical nonlinearity.
3rd Seminar
Date : 22nd March 2016
The program was as follows:
Speaker: Mr. Thomas Hu, CTO /VP, Magic Display Technology Co. Ltd.
Topic: present and prospect of the glasses-free 3D display technology
Abstract: Compared to the 2D display technology the multi-view 3D display technology allows the viewer to have different image of the observed object or sciences in different position space. This technology has great chance to be applied in the field of medical, teaching, television broadcasting. In this report the speaker will introduce the history and also the opportunity in the recent years of the 3D display technology. Basic technical introduction and the technical elements which play the key-roles will also be explained. Magic Display Technology Co. Ltd has the headquarter in Shenzhen. It has the international leading 3D technology and volume production lines for the glasses-free displays in the market.
Speaker: Mr. Meiping Li, Senior Vice President Yuan Guang Software Co. Ltd.
Topic: IT引领企业管理变革
Abstract: 随着云计算、大数据、物联网、移动应用的逐步普及，互联网新技术已经给我们生活、工作等方面带来了翻天覆地的变化。现今新技术的推动使得信息化在企业管理中的地位也在发生着变化，原来的IT支撑部门逐步转变为引领企业变革的核心部门。那么新的IT趋势是什么？企业管理的信息化正在发生什么样的变化？云大物移新技术如何引领企业管理的变革？本次报告带给大家一些个人思考，纯属抛砖引玉。
Speaker: Mr. Denis, Chan Siu Kee, Senior Manager, Information Systems Department, Companhia de Electricidade de Macau - CEM, S.A.
Topic: IT systems in CEM Macau
Abstract: CEM is the only utility in Macau supplying electricity to over 240,000 customers. As a company providing critical infrastructure to Macau SAR, IT is being used heavily to support the business to achieve a reliable and quality supply of our services. This presentation will give a general introduction to the various IT systems in CEM including examples such as ERP, Geographic Information System (GIS), Data analytic tools as well as other control and monitoring systems to supervise the electric network activities.