IEEE Tutorial on Deep Learning
IEEE Workshop on Deep Learning
Distributed Artificial Intelligence Laboratory for Interdisciplinary Research (2020-22)

Dist-AI Lab: Message

AI is penetrating into many aspects of human life so that both research in AI as well as in interdisciplinary areas with AI are much needed. We need to create enough critical mass in AI with an environment that provides easier path to learn and to conduct research, and the establishment of the distributed AI laboratory is a start of creating such an environment in 2020.

As more people in the Institute are conducting research with AI, more related research projects will be realized. While the CIS staff are applying for other AI related research projects, collaboration with other schools in other new projects have also started.

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: Objective

The objectives are:

  1. to bootstrap competence in AI technology research (which in turn benefits the curriculum) with the initial laboratory and learning environment,
  2. to enable participants to learn AI, to begin AI research, which can lead to further AI research with follow-up research fund applications,
  3. to apply AI in other areas (outside computing and information sciences) which can lead to interdisciplinary research with collaboration among different schools.

Please also check the Vision and Mission of the School of CIS.

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: Project team

Prof. H. Anthony CHAN

Prof. Wan-Chi SIU

Dr. Xueting Tina LIU

Dr. Chun Pong Jacky CHAN

Dr. Ricky Kwok Cheong AU, School of Social Sciences

Dr. Patrick Chi Wai LEE, School of Humanities and Languages

Dr. Charles Kin Man CHOW, Scholl of Business and Hospitality Management

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: Research facilities

The AI facility consists of:

  1. servers with arrays of the GPU's to conduct the most GPU-intensive deep learning algorithms and modeling
  2. servers for data storage to enable building up in-house data, especially in the kind of data that are not available in the public clouds
  3. personal AI machines for individuals to experiment with quick deep learning and AI tasks

Research with in-house AI facility is therefore supported both with high power AI servers or the average daily tasks.

Some interfaces are also being established to support interdisciplinary research, i.e., research traditionally in non-computer science may now involve AI.

Some IOT with network cards are being established with distributed AI in network research.

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: Contact Us

Please contact the School of Computing and Information Sciences or visit the School's office

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: Research Seminars

29 Nov 2019 Enhancing the Way to Secure Research Fund from UGC Prof. Wan-Chi SIU
20 February 2020 Neural Network & BP Algorithm Dr. Yingchao ZHAO
12 March 2020 Non-linear Function and Filter Structure in CNN Prof. Wan-Chi SIU
2 April 2020 Distributed and Federated Learning Prof. H. Anthony CHAN
16 April 2020 AlexNet – A Standard Structure of CNN Prof. Wan-Chi SIU
4 May 2020 End-to-end CNN and Image Segmentation Dr. Xueting LIU
21 May 2020 ResNet – A Breakthrough in Deep Learning Prof. Wan-Chi SIU
19 June 2020 Reinforcement learning basis and application case in networking: Augmenting Drive-thru Internet via Reinforcement Learning based Rate Dr. Wenchao XU
2 July 2020 Interdisciplinary research to prepare for the Age of Artificial Intelligence Prof. H. Anthony CHAN
6 July 2020 GAN – Generative Adversarial Network Prof. Wan-Chi SIU
20 July 2020 GAN (Part II) – Training and More Applications Prof. Wan-Chi SIU
11 August 2020 Introduction to AI in Natural Language Processing Prof. H. Anthony CHAN
25 August 2020 Lab Highlight and Region-based Convolutional Neural Network Prof. Wan-Chi SIU
8 September 2020 More RCNN and Additions to Lab Highlight Prof. Wan-Chi SIU
19 Nov 2020 Reinforcement learning demo: Q learning and deep Q learning Dr. Wenchao XU
8 December 2020 Baseline Model Design with Joint Back Projection and Residual Network Prof. Wan-Chi SIU
15 December 2020 Baseline Model Design with Joint Back Projection and Residual Network (Part II) Prof. Wan-Chi SIU

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: IEEE Joint Tutorial on Deep Learning

A tutorial on Deep Learning is initiated and is planned to be organized in each of the 3 years with this IDS fund. It is organizing jointly with IEEE a 2-day tutorial and is public to enable participants to learn or to deepen understanding of deep learning. Optional laboratory sessions to provide hands-on practice are also provided.

The tutorial in the year 2020 is postponed owing to pandemic to 5-6 March 2021. The announcement is being publicized at 2-Days Tutorial on Deep Learning and is also announced through other professional organizations.

Return to Dist-AI Lab Menu

Return to top

Dist-AI Lab: IEEE Joint Workshop on Deep Learning

A workshop on Deep Learning is initiated and is planned to be organized in each of the 3 years with this IDS fund. It is organized jointly with IEEE a 2-day workshop on continuous education.

The workshop in the year 2020 is postponed owing to pandemic to 19-20 March 2021. The announcement is being publicized at 2-Days Workshop on Deep Learning and is also announced through other professional organizations.

Return to Dist-AI Lab Menu

Return to top

Interdisciplinary Artificial Intelligence Research

Image Processing
Distributed Machine Learning
Social Sciences AI research
Humanity and Language AI research
Business and Hospitality management AI research
Artificial Intelligence Research Team

Prof. H. Anthony CHAN

Prof. Wan-Chi SIU

Dr. Xueting Tina LIU

Dr. Chun Pong Jacky CHAN

Dr. Ricky Kwok Cheong AU

Dr. Patrick Chi Wai LEE

Dr. Charles Kin Man CHOW

Return to top

Artificial Intelligence Research

Computer and Information Sciences AI research

Computer/Information Sciences and Health Sciences AI research

Computer/Information Sciences and Social Sciences AI research

Social Sciences AI research

Humanity and Language AI research

Business and Hospitality management AI research

Return to top

Prof. H. Anthony CHAN, BSc (HKU), MPhil (CUHK), PhD (Maryland), FIEEE

Professor and Dean, School of Computing and Information Sciences

5G Wireless network, Distributed machine learning, IETF (IPv6 Internet) standards, Components/Systems and Products Reliability

Room A804-1; Tel: 3702 4210; hhchan at cihe.edu.hk

Distributed Artificial Intelligence Lab

CIHE is awarded Institutional Development Scheme (IDS) Research Infrastructure Grant:
Establishment of Distributed Artificial Intelligence Laboratory for Interdisciplinary Research at $2.8 Million for 1/1/2020 - 31/12/2022.

Team leader: Prof. CHAN Hing Hung Anthony

Team members:

Prof. SIU Wan Chi

Dr. LIU Xueting Tina

et al.

Facility

The AI facility consists of

  1. servers with arrays of the GPU's to conduct the most GPU-intensive deep learning algorithms and modeling
  2. servers for data storage to enable building up in-house data, especially in the kind of data that are not available in the public clouds
  3. personal AI machines for individuals to experiment with quick deep learning and AI tasks

Research with in-house AI facility is therefore supported both with high power AI servers or the average daily tasks.

Some interfaces are also being established to support interdisciplinary research, i.e., research traditionally in non-computer science may now involve AI.

Some IOT with network cards are being established with distributed AI in network research.

Distributed machine learning in 5G Wireless

The current/emerging 5G Wireless is not just providing much higher data rate and expected to have 100 billion Internet Of all Things (IOT) IP devices using 5G Wireless network by year 2025. A few important technologies are:

3GPP is also defining the architecture, functions, and interfaces to enable use of machine learning in 5/6G Wireless. Network data analytics function (NWDAF) can employ machine learning to unlock the value of network data in the 5G Wireless core network, whereas machine learning may also be employed in the radio access network (RAN).

In addition, the IOT devices provide a large amount of data to support learning whereas learning may also be distributed or centralized.

Return to top

Prof. Wan-Chi SIU, MPhil (CUHK), PhD (Imperial College London), IEEE Live-Fellow

Research Professor, School of Computing and Information Sciences

DSP and Fast Algorithms, Intelligent Image and Video Coding, HDTV and 3DTV, Super-Resolution with Machine Learning and Deep Learning, Visual Surveillance, Object Recognition and Tracking, Visual Technology for Vehicle Safety and Autonomous Car

Room A804; wcsiu at cihe.edu.hk

Football Commentary System

Prof. SIU is awarded Institutional Development Scheme (IDS) Collaborative Research Grant:
Creating an Automatic Football Commentary System with Image Recognition and Cantonese Voice Output at $7 Million for 1/1/2021 - 31/12/2023.

Project Coordinator (PC): Prof. SIU Wan Chi

Co-PI: Prof. CHAN Sin-wai

Co-I: Prof. CHAN Hing Hung

Return to top

Dr. Xueting Tina LIU, BEng, BA (Tsinghua), PhD (CUHK)

Assistant Professor, School of Computing and Information Sciences

Computer Graphics, Computer Vision, Machine Learning, Computational Manga and Anime

Room A806; Tel: 3702 4217; tliu at cihe.edu.hk

Return to top

Image Processing research

Deep Extraction of Manga Structural Lines

Given a pattern-rich manga, we proposed a novel method to identify the structural lines of the manga image, with no assumption on the patterns. To suit our purpose, we propose a deep network model to handle the large variety of screen patterns and raise output accuracy. We also develop an efficient and effective way to generate a rich set of training data pairs.

Return to top

Invertible Grayscale

In this work, we propose an innovative method to synthesize invertible grayscale. It is a grayscale image that can fully restore its original color. The key idea here is to encode the original color information into the synthesized grayscale, in a way that users cannot recognize any anomalies. We propose to learn and embed the color-encoding scheme via a convolutional neural network (CNN). We then design a loss function to ensure the trained network possesses three required properties: color invertibility, grayscale conformity, and resistance to quantization error.

Return to top

Deep Binocular Tone mapping

Tone mapping is a commonly used technique that maps the set of colors in high-dynamic-range (HDR) images to another set of colors in low-dynamic-range (LDR) images. Recently, with the increased use of stereoscopic devices, the notion of binocular tone mapping has been proposed. The key is to map an HDR image to two LDR images with different tone mapping parameters, one as left image and the other as right image, so that more human-perceivable visual content can be presented with the binocular LDR image pair than any single LDR image. In this work, we proposed the first binocular tone mapping operator to more effectively distribute visual content to an LDR pair, leveraging the great representability and interpretability of deep convolutional neural network. Based on the existing binocular perception models, novel loss functions are also proposed to optimize the output pairs in terms of local details, global contrast, content distribution, and binocular fusibility.

Return to top

Deep Visual Sharing with Colorblind

Visual sharing between color vision deficiency (CVD) and normal-vision audiences is challenging due to the need of simultaneous satisfaction of multiple binocular visual requirements, in order to offer a color-distinguishable and binocularly fusible visual experience to CVD audiences, without hurting the visual experience of the normal-vision audiences. In this paper, we propose the first deep-learning based solution for solving this visual sharing problem. To achieve this, we propose to formulate this binocular image generation problem as a generation problem of a difference image, which can effectively enforce the binocular constraints. We also propose to retain only high-quality training data and enrich the variety of training data via intentionally synthesizing various confusing color combinations.

Return to top

Colorblind-Shareable Videos by Synthesizing Temporal-Coherent Polynomial Coefficients

To share the same visual content between color vision deficiencies (CVD) and normal-vision people, attempts have been made to allocate the two visual experiences of binocular display (wearing and not wearing glasses) to CVD and normal-vision audiences. However, existing approaches only tailor for still images. In this paper, we propose the first practical solution for fast synthesis of temporal-coherent colorblind-sharable video. Our fast solution is accomplished by a convolutional neural network (CNN) approach. To avoid the color inconsistency, we propose to use a global color transformation formulation. It first decomposes the color channels of each input video frame into several basis images and then linearly recombines them with the corresponding coefficients. Besides the color consistency, our generated colorblind-sharable video also needs to satisfy four constraints, including the color distinguishability, the binocular fusibility, the color preservation, and temporal coherence. Instead of generating the left and right videos separately, we train our network to predict temporal-coherent coefficients for generating a single difference video (between left and right views), which in turn to generate the binocular pair.

Return to top

Return to top

Dr. Chun Pong Jacky CHAN, BSc, PhD (CityUHK)

Assistant Professor

Computer Graphics, Computer Vision, Machine Learning, Character Animation, Health Care Technology

Room A810; Tel: 3702 4204; j2chan at cihe.edu.hk

Return to top

Classification of Parkinsonian Versus Normal 3-D Reaching

An objective assessment for determining whether a person has Parkinson disease is proposed. This is achieved by analyzing the correlation between joint movements, since Parkinsonian patients often have trouble coordinating different joints in a movement. It provides classification of subjects as having or not having Parkinson’s disease using the least square support vector machine (LS-SVM). Experimental results showed that using either auto-correlation or cross-correlation features for classification provided over 91% correct classification.

Return to top

Improving posture classification accuracy for depth sensor-based human activity monitoring in smart environments

Smart environments and monitoring systems are popular research areas nowadays due to its potential to enhance the quality of life. Applications such as human behavior analysis and workspace ergonomics monitoring are automated, thereby improving well-being of individuals with minimal running cost. In this paper, we propose a framework that accurately classifies the nature of the 3D postures obtained by Kinect using a max-margin classifier. Different from previous work in the area, we integrate the information about the reliability of the tracked joints in order to enhance the accuracy and robustness of our framework. As a result, apart from general classifying activity of different movement context, our proposed method can classify the subtle differences between correctly performed and incorrectly performed movement in the same context. We demonstrate how our framework can be applied to evaluate the user’s posture and identify the postures that may result in musculoskeletal disorders. Such a system can be used in workplace such as offices and factories to reduce risk of injury. Due to the low cost and the easy deployment process of depth camera based motion sensors, our framework can be applied widely in home and office to facilitate smart environments.

Return to top

A generic framework for editing and synthesizing multimodal data with relative emotion strength

Emotion is considered to be a core element in performances. In computer animation, both body motions and facial expressions are two popular mediums for a character to express the emotion. However, there has been limited research in studying how to effectively synthesize these two types of character movements using different levels of emotion strength with intuitive control, which is difficult to be modeled effectively. In this work, we explore a common model that can be used to represent the emotion for the applications of body motions and facial expressions synthesis. Unlike previous work that encode emotions into discrete motion style descriptors, we propose a continuous control indicator called emotion strength by controlling which a data-driven approach is presented to synthesize motions with fine control over emotions. Our method can be applied to interactive applications such as computer games, image editing tools, and virtual reality applications, as well as offline applications such as animation and movie production.

Return to top

Dr. Ricky Kwok Cheong AU, BSocSc MPhil (HKU), PhD (Univ of Tokyo)

Assistant Professor, School of Social Sciences

Room A626; Tel: 3702 4528; rau at cihe.edu.hk

Return to top

Social Sciences and artificial intelligence research

Emotional expression recognition

The School of Social Sciences is investigating emotional expressions with facial recognition aiming to achieve higher accuracy which has been challenging the prior art to recognize emotions in the difficult scenarios.

Return to top

Dr. Patrick Chi Wai LEE, MBA (Leicester, UK), MA (HKBU), MA (CityUHK), PhD (Newcastle UK)

Assistant Professor, School of Humanities and Languages

Theoretical Linguistics (syntax), Second Language Acquisition, Discourse Analysis

Room A617; Tel: 3702 4303; cwlee at cihe.edu.hk

Return to top

Humanities and Languages AI research

Language and artificial intelligence research:

The School of Humanities and Languages is investigating the use of different responses or types of speech acts between interlocutors (particularly the workplace English for students) to understand the use of various responses or types of speech acts and their functions in a specific scenario.

Return to top

Dr. Charles Kin Man CHOW, BSc (CityUHK), MBA (HKPolyU), MSc PGCE (HKU), DBA (Newcastle, Australia)

Assistant Professor, Rita Tong Liu School of Business and Hospitality Management

Management information systems, e-commerce, and e-service quality

Room A511; Tel: 3702 4568; cchow at cihe.edu.hk

Return to top

Business and Hospitality management AI research

Smart hotel

Current work includes smart hotel, such as smart scenery, smart guest services. Other areas in accounting and finance are also being explored.

In a previous study (Chow, 2017), Dr. Charles Chow examined how web design, responsiveness, reliability, enjoyment, ease of use, security, and customization influence e-service quality of online hotel booking agencies in Hong Kong. It was found that these factors positively influenced e-service quality of online hotel booking agencies. The research provided new insights to industry practitioners with recommendations in designing and implementing online hotel booking websites. With these findings taking into consideration, online hotel booking companies can concentrate their resources to the identified dimensions, in particularly through web design and customization, to achieve the desired level of e-service quality.

Reference

Chow, K. M. (2017). E-service quality: A study of online hotel booking websites in Hong Kong. Asian Journal of Economics, Business and Accounting, 3(4), 1-13.

Return to top

AI References

References in AI

Introductory Guide to Artificial Intelligence

Machine learning resources

10 trends of AI in 2019

Artificial Intelligence Programmes

Deep Learning course at Univ Wisconsin-Madison

Introduction to Deep Learning (CUHK)

Deep Learning, by Ian Goodfellow and Yoshua Bengio and Aaron Courville

CS229: Machine Learning

Is LSTM Dead, Long Live Transformers

How do Transformers Work in NLP?

AI for Everyone(deeplearning.ai)

Deep Learning Specialization(deeplearning.ai)

Return to top

References in Healthcare

How Artificial Intelligence Helps in Health Care

Artificial Intelligence Will Change Healthcare as We Know It

AI In Health Care: The Top Ways AI Is Affecting The Health Care Industry

Surgical robots, new medicines and better care: 32 examples of AI in healthcare

HL7:to develop American National Standards Institute ANSI-accredited standards for the exchange, integration, sharing and retrieval of electronic health information that supports clinical practice and the management, delivery and evaluation of health services.

CCD:Continuity of Care Document is built using HL7 Clinical Document Architecture (CDA) elements and contains data that is defined by the American Society for Testing and Materials (ASTM) Continuity of Care Record (CCR) to share summary information about the patient within the broader context of the personal health record.

Jiang F, Jiang Y, Zhi H, et al., "Artificial intelligence in healthcare: past, present and future," Stroke and Vascular Neurology 2017;2:doi: 10.1136/svn-2017-000101

Dhamdhere P, Harmsen J, Hebbar R, et al., "Big Data in Healthcare," 2016.

Joyce C, "Harnessing the Power of Data in Health Care: Data as a Strategic Asset,"Symposium on Data Science for Healthcare (DaSH), 2017.

Ettinger A, "Data and Information Management in Public Health," July 2004.

AI for Medicine Specialization(deeplearning.ai)

Return to top

References in Management

The Place of Management in an AI Curriculum

Managing Human and Machine Intelligence

What does human-centric AI mean to management?

Business Analytics Institute

Return to top

References in Social services

Future of Social Work, GISW Anniversary Celebration, Nov 23 2018.

Future of Social Work: International Perspectives

How Artificial Intelligence Will Save Lives in the 21st Century

Artificial Intelligence Ideas and Social Work

Applying AI for social good

Notes from the ai frontier applying ai for social good

Envisioning the future of social work

Return to top

AI demo

AI experiments with Google

10 craziest AI experiments

Return to top

AI: Natural Language Processing

Deep Learning for NPL: An Overview of Recent Trends

Break into NLP

AI for Natural Language Processing Specialization(deeplearning.ai)

Return to top