Nectar Program

The Nectar program consists of Taiwan's latest AI research at top conferences, including AAAI, ACL, ACMMM, ECCV, ICML, KDD, NeurIPS, SIGIR and WWW. (More information ...)

Talk Title: Why Attention? Analyze BiLSTM Deficiency and Its Remedies in the Case of NER

Generic placeholder image
Speaker: Prof. Wei-Yun Ma
Associate Research Fellow, Institute of Information Science, Academia Sinica
Time: 10:30-11:00
Biography

Dr. Wei-Yun Ma is an associate research fellow/professor in the Institute of Information Science at Academia Sinica in Taiwan. He is also the principal investigator of the NLP lab at Academia Sinica, a.k.a CKIP lab. His research interests include NLP, machine learning, and knowledge graph.He is currently working on knowledge reasoning, Chatbot, data-to-text, multi-hop QA, and knowledge graph completion.He mainly publishes articles in international conferences such as AAAI, ACL, EMNLP, NAACL, SigDial, COLING, EACL, WWW, etc.Dr. Ma has also led the team to develop and release various popular Chinese NLP tools or resources, such as CkipTagger and CKIP CoreNLP.



Talk Title: HardGAN: A Haze-Aware Representation Distillation GAN for Single Image Dehazing

Generic placeholder image
Speaker: Prof. Chia-Wen Lin
Professor, National Tsing Hua University
Time: 11:00-11:30
Biography

Dr. Lin was named IEEE Fellow in 2018. He was a Distinguished Lecturer of IEEE Circuits and Systems Society (2018~2019). He has served on the editorial boards of IEEE Transactions on Image Processing, IEEE Transactions on Multimedia, and IEEE Transactions on Circuits and Systems for Video Technology, He was Chair of the Multimedia Systems and Applications Technical Committee of IEEE Circuits and Systems Society (2013~2015). He is Steering Committee Chair of IEEE ICME. He served as Lead TPC Chair for IEEE ICIP 2019 and IEEE ICME 2010. He also served as General Co-Chair of IEEE VCIP 2018.



Talk Title: Knowledge-Enriched Visual Storytelling

Generic placeholder image
Speaker: Prof. Lun-Wei Ku
Associate research fellow/ Academia SINICA, Adjunct associate professor/NCTU Secretary-general of Association for Computational Linguistics and Chinese Language Processing (ACLCLP).
Time: 14:50-15:20
Biography

Prof. Ku's research expertise lies in natural language processing, and information retrieval, especially in sentiment analysis and opinion mining. She is very active in the research community, and the international professional activities she involves include the program chair of StarSem 2019 and ARIS 2019, and the area chair of ACL, EMNLP, COLING 2020, EMNLP 2019, ACL 2017, CCL , NLPCC 2016, ACL-IJCNLP and EMNLP 2015. She is also very experienced in academic and industrial collaborations. Her research collaborators come from US, Singapore, Sweden and Israel, and she is currently working with data companies and banks. Her current research topics focus on recommendation, visual storytelling, sensational text generation, fake news intervention, knowledge-based question answering, lie detection and social media analysis.



Talk Title: On the Robustness of Self- Attentive Models

Generic placeholder image
Speaker: Dr. Yu-Lun Hsieh
Postdoctoral Fellow, Institute of Information Science, Academia Sinica
Time: 15:20-15:50
Biography

Yu-Lun Hsieh received his Ph.D. from TIGP, Academia Sinica and National Chengchi University. Prior to that, he obtained an M.A. in Linguistics and B.Sc. in Computer Science from National Tsing-Hua University. His research interests include machine learning, natural language processing, deep neural models, adversarial attacks, and computational linguistics. In 2018, he was invited as a visiting scholar at University of California, Davis, in collaboration with Google AI, to explore adversarial robustness of neural network models.



Talk Title: Attractive or Faithful? Popularity-Reinforced Learning for Inspired Headline Generation

Generic placeholder image
Speaker: Ms. Yun-Zhu Song
Ph.D. Student, National Chiao Tung University
Time: 15:50-16:20
Biography

Hello, my name is Yun-Zhu Song (Angela Song).I am currently a second-year Ph.D. student from Nation Chiao Tung University and major in deep learning for Natural Language Generation. I am interested in controlling the text characteristic and aim to find a better solution for feature manipulation. Additionally, due to the growing impact of social media, the behavior of rumor and fake news is also one of my research targets.



Talk Title: Compacting, Picking and Growing for Unforgetting Continual Learning

Generic placeholder image
Speaker: Prof. Cheng-Hao Tu
Research Assistant, Academia Sinica
Time: 10:50-11:20
Biography

He is currently a research assistant at Institute of Information Science, Academia Sinica. He received his B.Sc. degrees at the Department of Computer Science & Information Engineering, National Taiwan University. His research interest includes computer vision, hash code learning, continual learning and network pruning.



Talk Title: Accelerating and Improving AlphaZero Using Population Based Training

Generic placeholder image
Speaker: Dr. Ti-Rong Wu
Post-Doctoral Fellow, National Chiao Tung University
Time: 11:20-11:50
Biography

He received his Ph.D. degree in Computer Science at National Chiao Tung University in 2020. His research interests include computer games, artificial intelligence, reinforcement learning, deep learning, and machine learning. He led a group that developed a computer Go program, named CGI Go Intelligence, which won second place in the First World AI Go Open in 2017.



Talk Title: Personal Knowledge Base Construction from Text-based Lifelogs

Generic placeholder image
Speaker: Dr. An-Zi Yen
Post-Doctoral Fellow, National Taiwan University
Time: 13:30-14:00
Biography

Dr. An-Zi Yen received the Ph.D. degree in computer science and information engineering in 2020 from National Taiwan University, Taipei, Taiwan. She is currently a Post-Doctoral Fellow in the Department of Computer Science and Information Engineering at the National Taiwan University. Her research interests include natural language processing and information retrieval. Her work has been published in SIGIR, WWW, WI, ICMR, and so on.



Talk Title: Probing Neural Network Comprehension of Natural Language Arguments

Generic placeholder image
Speaker: Mr. Timothy Niven
Research Scientist, National Cheng Kung University
Time: 14:00-14:30
Biography

I am originally a philosophy graduate and came to computational linguistics later in life. I am still a PhD student at the Intelligent Knowledge Management Lab at National Cheng Kung University, under professor Kao Hung-Yu. I have recently commenced working as a research scientist at Doublethink Lab in Taipei, working on information warfare.



Talk Title: Efficient Algorithms towards Network Intervention

Generic placeholder image
Speaker: Ms. Hui-Ju Hung
Ph.D Candidate, The Pennsylvania State University
Time: 14:30-15:00
Biography

Currently a Ph.D. candidate in the Pennsylvania State University. She received the B.S. degree from National Tsing Hua University and the M.S. degree from National Taiwan University. Her research interests include optimization and analysis on social networks. Some of her work has been published in ACM KDD, ACM CIKM, IEEE ICDCS, IEEE GLOBECOM, and IEEE TMC.



Talk Title: Conquering Cross-source Failure for News Credibility: Learning Generalizable Representations beyond Content Embedding

Generic placeholder image
Speaker: Mr. Yen-Hao Huang
Ph.D. Candidate, National Tsing Hua University
Time: 15:10-15:40
Biography

He is a Ph.D. Candidate and Teacher Assistant with the Institute of Information Systems and Applications, National Tsing Hua University since 2015. His current research interests include fake news analysis, emotion recognition, computational mental health, natural language processing, ,deep learning, computational mental health analysis



Talk Title: FashionOn: Semantic-guided Image- based Virtual Try-on with Detailed Human and Clothing Information

Generic placeholder image
Speaker: Prof. Wen-Huang Cheng
Prof. Cheng (IET Fellow) is Director of NCTU AI Graduate Program and Scientific Advisor to Taipei City Government. He is Associate Editor for IEEE Transactions on Multimedia, General co-chair for ACM ICMR (2021), TPC co-chair for ICME (2020), Chair-Elect for IEEE MSA-TC, governing board member for IAPR.
Time: 15:40-16:10
Biography

His research interest includes multimedia, artificial intelligence, computer vision, machine learning, social media, and financial technology. He received research honors including 2018 MSRA Collaborative Research Award.



Talk Title: DATE: Dual Attentive Tree-aware Embedding for Customs Fraud Detection

Generic placeholder image
Speaker: Mr. Yu-Che Tsai
Master student, National Taiwan University
Time: 10:00-10:30
Biography

Bachelor’s degree in Statistical Science from NCKU, Tainan, Taiwan,in 2020. He is also a research assistant in Networked Artificial Intelligence Laboratory at NCKU. Main research interests include Data Mining, Machine Learning, Recommender Systems, and Deep Learning. His papers had been published in KDD 2020, CIKM 2019, and RecSys 2019.



Talk Title: Weakly Supervised Instance Segmentation using the Bounding Box Tightness Prior

Generic placeholder image
Speaker: Mr. Cheng-Chun Hsu
Research Scientist (Fixed-Term),Pro Unlimited @ Facebook
Time: 10:30-11:00
Biography

I have worked as a research scientist in Facebook Reality Lab at Pro Unlimited @ Facebook since 2020. Previously, I was a research assistant in Vision and Learning Lab at Academia Sinica, advised by Prof. Yen-Yu Lin. My research focuses on computer vision and perception of machines. Particularly, my work aims to enable the machines to learn to understand visual data and improve its recognition ability with less human supervision. I am interested in Domain Adaptation, Instance Segmentation, and Weakly Supervised Learning.



Talk Title: Weakly-Supervised Video Re- Localization with Multiscale Attention Model

Generic placeholder image
Speaker: Mr. Yung-Han Huang
Machine Learning Engineer , Perfect Corp.
Time: 11:00-11:30
Biography

I graduated from National Taiwan University with a master's degree in data science. I am interested in computer vision and deep learning. Specifically, my master thesis was focused on video related topics and weakly supervised methods.



Talk Title: Transfer Learning without Knowing: Reprogramming Black-box Machine Learning Models with Scarce Data and Limited Resources

Generic placeholder image
Ms. Yun-Yun Tsai
Research assistant, National Tsing Hua University
Time: 11:30-12:00
Biography

Yun-Yun Tsai is currently a research assistant in the Department of Computer Science of National Tsing Hua University (NTHU). She has been working on artificial intelligence in security and the robustness in deep learning, including adversarial attack, defense, and robust evaluation. She has several paper publications in top conferences, such as ICML, NeurIPS workshop, and NDSS. Yun-Yun earned M.S. and B.S. in Computer Science in 2020 and 2018, both from NTHU. Currently, she worked with Professor Tsung-Yi Ho and Dr. Pin-Yu Chen from IBM Research.