IDS Guest Seminar: Visual Perception and Learning in an Open World

Title: Visual Perception and Learning in an Open World
Speaker: Dr Shu KONG, Assistant Professor, Department of Computer Science and Engineering, Texas A&M University
Date: May 23, 2023
Time: 10:00am – 11:00am

Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.

Abstract

Visual perception is indispensable in numerous applications spanning autonomous vehicles and interdisciplinary research. Today’s visual perception algorithms are often developed under a closed-world paradigm, which assumes the data distribution and categorical labels are fixed a priori. This assumption is unrealistic in the real open world, which contains vast situations that are unpredictable and dynamic. As a result, closed-world visual perception systems appear to be brittle in the open-world. For example, equipped with such visual perception systems, an autonomous vehicle could fail to recognize a never-before-seen overturned truck and cause collision; it could fail to detect a pedestrian in a dark night and cause casualties. In this talk, I will present my solutions to recognizing unknown objects, segmenting and detecting general objects, and improving object detection using multimodal signals. If time allows, I will share my thought about open-world visual perception and learning, and sketch my future research.

Speaker

Dr Shu KONG
Assistant Professor @ Department of Computer Science and Engineering, Texas A&M University
Shu Kong is on the faculty in the Department of Computer Science and Engineering, Texas A&M University, after postdoc training in the Robotics Institute, Carnegie Mellon University. He received the PhD degree in computer science from the University of California, Irvine. His research interests include computer vision, applied machine learning, and their broad applications. His current research focus is on visual perception and learning in the open world. His recent paper on this topic received honorable mention for Best Paper / Marr Prize at ICCV 2021. His latest interdisciplinary research develops high-throughput pollen analysis tools, which were featured by the National Science Foundation as that “open new era of fossil pollen research.”
Dr Su’s Google Scholar page: https://scholar.google.com/citations?user=sm9FdLoAAAAJ&hl=en

For information, please contact:
Email: datascience@hku.hk

IDS Seminar: Modeling the 3D Physical World for Embodied Intelligence

Title: Modeling the 3D Physical World for Embodied Intelligence
Speaker: Dr Hao Su, Assistant Professor, Department of Computer Science and Engineering, Jacobs School of Engineering,
University of California, San Diego
Moderator: Professor Kenneth Wong, Head of Department of Electrical and Electronic Engineering, HKU
Date: May 9, 2023
Time: 10:30am – 11:30am

Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.

Seminar recording:

Abstract

Embodied AI is a rising paradigm of AI that targets enabling agents to interact with the physical world. Embodied agents can acquire a large amount of data through interaction with the physical world, which makes it possible for agents to close the perception-cognition-action loop and learn autonomously from the world to revise its internal model. In this talk, I will present the work of my group to build an eco-system towards building embodied AI.

Speaker

Dr Hao Su
Assistant Professor @ Department of Computer Science and Engineering, University of California, San Diego
Hao Su is an Assistant Professor of Computer Science at the University of California, San Diego. He is the Director of the Embodied AI Lab at UCSD. He works on algorithms to model, undertand, and interact with the physical world. His interests span computer vision, machine learning, computer graphics, and robotics – all areas in which he has published and lectured extensively. He has more than 75,000 citations according to Google Scholar. Hao Su obtained his Ph.D. from Stanford in 2018. At Stanford and UCSD he developed widely used datasets and softwares such as ImageNet, ShapeNet, PointNet, PartNet, SAPIEN, and more recently, ManiSkill. He also developed new courses to promote machine learning methods for 3D geometry and embodied AI. He served as the Area Chair or Associate Editor for top conferences and journals in computer vision (ICCV/ECCV/CVPR), computer graphics (SIGGRAPH/ToG), robotics (IROS/ICRA), and machine learning (ICLR/NeurIPS).
Dr Su’s Google Scholar page: https://scholar.google.com/citations?user=1P8Zu04AAAAJ&hl=en

Moderator

Prof Kenneth K.Y. Wong
Head @ Department of Electrical and Electronic Engineering, HKU
Prof. Kenneth Kin-Yip Wong is currently a Professor in the Department of Electrical and Electronic Engineering in the University of Hong Kong. He is a senior member of the IEEE (Photonics Society), OSA, and SPIE. He received combined B.E. (1st class honor with medal award) degree in electrical engineering and B. S. degree in physics from the University of Queensland, Brisbane, Australia, in 1997. He received the M.S. degree in 1998 and the Ph.D. degree in 2003, both in electrical engineering at Stanford University. He was a member of the Photonics and Networking Research Laboratory at Stanford University. His research field included DWDM systems, SCM optical systems, fiber nonlinearity, and fiber optical parametric amplifiers. He is author or coauthor of over 50 journal and conference papers.
For full biography of Prof. Wong, please browse his website at: https://www.eee.hku.hk/people/kywong/

For information, please contact:
Email: datascience@hku.hk

Research team led by HKU IDS Scholar Dr Chao Huang ranked 1st & 3rd in the Most Influential Papers of SIGIR 2022 (2023-04 ver.)!

Research team led by HKU IDS Scholar Dr Chao Huang ranked 1st and 3rd in the Most Influential Papers of SIGIR 2022 (2023-04 ver.)!

Huge congratulations to our HKU IDS scholar, Dr Chao Huang, and his research group for being featured in the Most Influential SIGIR Papers (2023-04 ver.). Two research papers under Dr Huang’s supervision, namely “Hypergraph Contrastive Collaborative Filtering” and “Knowledge Graph Contrastive Learning for Recommendation” are ranked 1st and 3rd respectively!

SIGIR (Annual International ACM SIGIR Conference on Research and Development in Information Retrieval) is one of the top-tier data mining and information retrieval conferences in the world. Please refer to the website below for the paper abstracts:
https://www.paperdigest.org/2023/04/most-influential-sigir-papers-2023-04/

Dr Chao Huang’s research interests cover data mining and machine learning. For more information about Dr Huang, please browse: https://datascience.hku.hk/people/chao-huang/

Accounting Executive (at the rank of Clerk II)

Careers at IDS

Accounting Executive (at the rank of Clerk II)

Position  

Applications are invited for the appointment as Accounting Executive (at the rank of Clerk II)
in the HKU Musketeers Foundation Institute of Data Science (HKU IDS) (Job Ref.: 522723), (to commence as soon as possible, on a two-year fixed-term basis with contract-end gratuity and University contribution to a retirement benefits scheme, totalling up to 10% of basic salary, with the possibility of renewal subject to satisfactory performance).

Applicants should possess a Higher Diploma, OR 5 passes in HKCEE including English (min. Grade C if Syllabus A/Level 2 from 2007), Chinese (Level 2 from 2007) and Mathematics, OR have obtained a minimum of Level 2 or equivalent in 5 subjects in HKDSEE including English Language, Chinese Language and Mathematics, with at least 3 years’ accounting experience, preferably in tertiary institutions or in the education sector and/or auditing.  They should have a good command of written and spoken English and Chinese, good interpersonal, communication and organisational skills, computer literacy, and ability to multitask under pressure and work independently as well as in a team.  They should also be detail-oriented and self-motivated with a strong sense of responsibility.  Those with strong knowledge of PC and software applications such as MS Word, Excel and PowerPoint, as well as experience with Oracle Financials, are highly preferred. 

The appointee will assist in financial and administrative matters of the Institute, including budgetary control, payments and receipts processing, preparation of accounting statements, staff reimbursement and outside practice, maintenance of database for finance records, and office inventory.  He/She will also provide clerical support for tasks as assigned by supervisor. 

Shortlisted candidates will be invited to attend a written test and an interview.  

A highly competitive salary commensurate with qualifications and experience will be offered, in addition to annual leave and medical benefits.

Application

The University only accepts online application for the above post.  Applicants should apply online and upload an up-to-date C.V.  Review of applications will commence as soon as possible and continue until September 30, 2023, or until the post is filled, whichever is earlier.

Apply Now

HKU IDS Scholar Dr Tao Yu received the 2023 Google Research Scholar Award!

HKU IDS Scholar Dr Tao Yu received the 2023 Google Research Scholar Award!

The Institute is proud to announce that our HKU IDS scholar, Dr. Tao Yu, Assistant Professor in the HKU IDS and Department of Computer Science, has been awarded the 2023 Google Research Scholar Award!

Dr. Yu’s award-winning proposal, under the category of “Structured data, extraction, semantic graph, and database management”, is titled “Building Natural Language Interfaces for Data Science with Language Models“. His research interests cover Natural Language Processing and Artificial Intelligence. For more information about Dr. Yu, please browse: https://datascience.hku.hk/people/tao-yu/

The Google Research Scholar Program aims to support early-career engineers and scientists who are in pursuit of world-class research projects. Congratulations to Dr. Yu on his outstanding achievement!

HKU IDS Scholar Seminar Series #5: Denoising Diffusion-based Generative Modes for Visual Content Generation

Title: Denoising Diffusion-based Generative Modes for Visual Content Generation
Speaker: Dr. Xihui Liu, Assistant Professor, IDS & Department of Electrical and Electronic Engineering, HKU
Date: May 5, 2023
Time: 10:30am – 11:30am

Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.

Seminar recording:

Abstract

Denoising diffusion models, also known as score-based generative models, are a class of generative models that can produce high-fidelity images, videos and 3D models from noise. They define a forward diffusion process that gradually adds noise to the input data, and a reverse diffusion process that iteratively denoises the noise using a learned score function. In this talk, we will introduce the foundations and applications of denoising diffusion-based generative models. We will cover the theoretical background, the training and sampling methods, the recent advances and challenges, and the practical use cases of these models. We will also demonstrate some examples of image/video/3D generation using denoising diffusion-based generative models and discuss their advantages and limitations compared to other generative models.

Speaker

Dr. Xihui Liu
Assistant Professor @ HKU IDS & Dept of EEE
Dr. Xihui Liu is an Assistant Professor at the Musketeers Foundation Institute of Data Science (HKU IDS) and the Department of Electrical and Electronic Engineering (EEE)The University of Hong Kong. Before joining HKU, she was a postdoc Scholar at UC Berkeley, advised by Prof. Trevor Darrell. She obtained her Ph.D. degree from Multimedia Lab (MMLab), the Chinese University of Hong Kong, supervised by Prof. Xiaogang Wang and Prof. Hongsheng Li. She received her bachelor’s degree in Electronic Engineering at Tsinghua University.
Her research interests cover computer vision, machine learning, and artificial intelligence, with special emphasis on visual synthesis, generative models, vision and language, and multimodal AI. She was awarded Adobe Research Fellowship 2020, MIT EECS Rising Stars 2021, and WAIC Rising Stars Award 2022.

Moderator

Prof Kenneth K.Y. Wong
Head @ Department of Electrical and Electronic Engineering, HKU
Prof. Kenneth Kin-Yip Wong is currently a Professor in the Department of Electrical and Electronic Engineering in the University of Hong Kong. He is a senior member of the IEEE (Photonics Society), OSA, and SPIE. He received combined B.E. (1st class honor with medal award) degree in electrical engineering and B. S. degree in physics from the University of Queensland, Brisbane, Australia, in 1997. He received the M.S. degree in 1998 and the Ph.D. degree in 2003, both in electrical engineering at Stanford University. He was a member of the Photonics and Networking Research Laboratory at Stanford University. His research field included DWDM systems, SCM optical systems, fiber nonlinearity, and fiber optical parametric amplifiers. He is author or coauthor of over 50 journal and conference papers.
For full biography of Prof. Wong, please browse his website at: https://www.eee.hku.hk/people/kywong/

For information, please contact:
Email: datascience@hku.hk

IDS Interdisciplinary Seminar Series: The Philosophy of Artificial Intelligence: What it Is, Why it Matters, and How it Can Influence the Development of AI

Host: HKU Musketeers Foundation Institute of Data Science
Co-host: AI & Humanity Lab, Department of Philosophy, HKU

IDS Interdisciplinary Seminar - by Professor Herman Cappelen

Title: The Philosophy of Artificial Intelligence: What it Is, Why it Matters, and How it Can Influence the Development of AI 
Speaker: Professor Herman Cappelen, Chair Professor of Philosophy; Director, AI & Humanity Lab, The University of Hong Kong
School of Humanities, The University of Hong Kong
Date: Apr 26, 2023
Time: 3:30pm – 4:30pm

Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.

Seminar recording:

Abstract

There’s broad agreement that the rapid development of artificial intelligence raises fundamental and existential challenges for humanity. This talk outlines the ways in which this development also raises challenging philosophical questions, why those questions matter, and how computer scientists and philosophers can collaborate on solving them.

Speaker

Professor Herman Cappelen
Chair Professor of Philosophy, School of Humanities; Director of AI & Humanity Lab @ The University of Hong Kong

Prof. Cappelen is Chair Professor to the Department of Philosophy at the University of Hong Kong and the Director of AI&Humanity-Lab@HKU. He is also serving in the Steering Committee of the HKU Musketeers Foundation Institute of Data Science.

Before coming to HKU, he was on the faculty at the University of Oxford, the University of St Andrews and the University of Oslo. Cappelen is the author of 10 monographs and over 50 papers, covering many areas of philosophy. Cappelen and Josh Dever wrote the first philosophy book on explainable and interpretable AI:  Making AI Intelligible: Philosophical Perspectives (Oxford University Press 2021). With Rachel Sterken (HKU), he is the editor of Communicating with AI: Philosophical Perspectives (forthcoming Oxford University Press 2023). Cappelen and Dever’s new book in progress is called: In Defence of Artificial Intelligences: An Essay in Inhuman Philosophy. Cappelen is an elected member of Academia Europaea and the Norwegian Academy of Arts and Sciences.  

For Prof. Herman’s full biography, please browse: https://datascience.hku.hk/professor-cappelen-herman/ 

Moderator

Dr. Janet Hsiao
Head and Associate Professor, Department of Psychology; Affiliate of AI & Humanity Lab @ The University of Hong Kong

Dr. Janet Hsiao is an associate professor in the Department of Psychology and a principal investigator of the State Key Laboratory of Brain and Cognitive Sciences at University of Hong Kong. She received her Ph.D. in Informatics from University of Edinburgh and was a postdoctoral researcher in the Temporal Dynamics of Learning Center at University of California San Diego. She is also serving in the Steering Committee of the HKU Musketeers Foundation Institute of Data Science.

As a cognitive scientist, she is best known for her research on learning and visual cognition. She investigates universal principles and specific factors that modulate development of perceptual representations and information processing strategies during learning and expertise acquisition such as face recognition, reading, and object detection and identification. She adopts an interdisciplinary approach, using a variety of methods and techniques from artificial intelligence, experimental psychology, psycholinguistics, and cognitive neuroscience to study the human mind at different levels of analysis and organization. Her unique interdisciplinary approach has deepened and broadened our understanding of individual differences in learning and visual cognition and how they connect to cognitive performance and disorders.

For Dr. Hsiao’s full biography, please browse: https://datascience.hku.hk/dr-hsiao-janet/

For information, please contact:
Email: datascience@hku.hk

IDS Guest Lecture: Guided Text-based Item Exploration

IDS Guest Lecture: by Professor Sihem Amer-Yahia

Title: Guided Text-based Item Exploration
Speaker: Professor Sihem Amer-Yahia, CNRS Research Director, Lab of Informatics of Grenoble, France
Moderator: Professor Reynold Cheng, Associate Director of HKU IDS; Professor, Department of Computer Science, HKU
Date: Apr 20, 2023
Time: 11:00am – 12:00pm

Mode: Face-to-face (Advanced registration is required for on-site participants, and a confirmation email will be sent to participants who have successfully registered.)

Register Now

Abstract

Exploratory Data Analysis (EDA) provides guidance to users to help them refine their needs and find items of interest in large volumes of structured data. I will present GUIDES, a framework for guided Text-based Item Exploration (TIE). TIE raises new challenges: (i) the need to abstract and query textual data and (ii) the need to combine queries on both structured and unstructured content. GUIDES represents text dimensions such as sentiment and topics, and introduces new text-based operators that are seamlessly integrated with traditional EDA operators. To train TIE policies, it relies on a multi-reward function that captures different textual dimensions, and extends the Deep Q-Networks (DQN) architecture with multi-objective optimization. Our experiments on Amazon and IMDb, two real-world datasets, demonstrate the necessity of capturing fine-grained text dimensions, the superiority of using both text-based and attribute-based operators over attribute-based operators only, and the need for multi-objective optimization.

Speaker

Professor Sihem Amer-Yahia
CNRS Research Director @ Lab of Informatics of Grenoble, France
Sihem Amer-Yahia is a Silver Medal CNRS Research Director and Deputy Director of the Lab of Informatics of Grenoble. She works on exploratory data analysis and fairness in job marketplaces. Before joining CNRS, she was Principal Scientist at QCRI, Senior Scientist at Yahoo! Research and Member of Technical Staff at at&t Labs. Sihem is PC chair for SIGMOD 2023 and vice president of the VLDB Endowment. She currently leads the Diversity, Equity and Inclusion initiative for the database community. 
Professor Amer-Yahia is a Visiting Research Professor, under the Visiting Research Professors Scheme for the year 2020-23 hosted by Department of Computer Science, The University of Hong Kong.

Moderator

Professor Reynold Cheng
Associate Director @ the Musketeers Foundation Institute of Data Science; Professor @ Department of Computer Science, HKU

Professor Reynold Cheng is a Professor of the Department of Computer Science in the University of Hong Kong (HKU). His research interests are in data science, big graph analytics and uncertain data management. He was the Assistant Professor in the Department of Computing of the Hong Kong Polytechnic University (HKPU) from 2005 to 2008. He received his BEng (Computer Engineering) in 1998, and MPhil (Computer Science and Information Systems) in 2000 from HKU. He then obtained his MSc and PhD degrees from Department of Computer Science of Purdue University in 2003 and 2005.

Prof. Cheng has received numerous academic awards, and he is also the Associate Director of the Musketeers Foundation Institute of Data Science. For full biography of Prof. Cheng, please browse: https://datascience.hku.hk/reynold-cheng/

For information, please contact:
Email: datascience@hku.hk

HKU IDS Scholar Seminar Series #4: Over-parameterization in Deep Learning: Kernel Regime and Beyond

Title: Over-parameterization in Deep Learning: Kernel Regime and Beyond
Speaker: Dr. Difan Zou, Assistant Professor, IDS & Department of Computer Science, HKU
Date: Apr 12, 2023
Time: 3:30pm – 4:30pm

Mode: Hybrid. Seats for on-site participants are limited. A confirmation email will be sent to participants who have successfully registered.

Seminar recording:

Abstract

In recent years, deep learning has revolutionized the field of artificial intelligence, achieving remarkable success in a variety of applications. However, the high-dimensional and nonconvex nature of deep neural networks has made it challenging to understand their behavior and performance. Over-parameterization, which refers to the phenomenon where neural networks have more parameters than necessary to fit the training data, has become a key concept in the study of deep learning. In this talk, the speaker will explore the concept of over-parameterization and introduce a series of recent works that fall in the so-called “kernel regime where the neural network behaves like a kernel method. Furthermore,I will discuss the advantages and limitations of these kernel-based analyses, and introduce several remarkable attempts beyond the kernel regime. Specifically, the speaker will discuss how over-parameterization can affect generalization, optimization, and sample complexity. Overall, this talk aims to provide a comprehensive overview of over-parameterization in deep learning, and to highlight key questions as well as further research directions in this exciting and rapidly evolving area.

Speaker

Dr. Difan Zou
Assistant Professor @ HKU IDS & Dept of CS
Dr. Difan Zou is an Assistant Professor in the HKU IDS and Department of Computer Science at the University of Hong Kong. He received his Ph.D. in Computer Science, University of California, Los Angeles (UCLA). He received a B. S degree in Applied Physics, from School of Gifted Young, USTC and a M. S degree in Electrical Engineering from USTC. He has published multiple papers on top-tier machine learning conferences including ICML, NeurIPS, ICLR, COLT, etc. He is a recipient of Bloomberg Data Science Ph.D. fellowship.
His research interests are broadly in machine learning, optimization, and learning structured data (e.g., time-series or graph data), with a focus on theoretical understanding of the optimization and generalization in deep learning problems.

For information, please contact:
Email: datascience@hku.hk

IDS Research Seed Funds 2023 – Now Open for Application!

IDS Research Seed Funds 2023 Invites Applications Until April 30, 2023

The application for the IDS Research Seed Funds 2023 (IDS-RSF 2023) is now open!

The Steering Committee of the Institute has supported the calling for funding exercise to encourage submissions for excellent proposals in specific areas of data science. We are now inviting submissions for excellent proposals in specific areas of data science, with the aim of achieving greater research impact and deliverables that will benefit the society and people around the world.

For interested eligible HKU members, please submit an electronic application, on or before April 30, 2023, by browsing the funding call circular below:

Learn more