He served as a program co-chair for the ACM Symposium on User Interface Software and Technology (UIST) in 2013, a conference co-chair for ACM UIST in 2016, technical papers chair for SIGGRAPH ASIA in 2018 and technical program co-chair for ACM CHI in 2021.ġ0 - Monday - 4:10 p.m. Igarashi has received several awards, including the Association for Computing Machinery's (ACM) SIGGRAPH 2006 Significant New Researcher Award, the ACM CHI Academy award 2018 and the Asia Graphics 2020 Outstanding Technical Contributions Award. His research interests are in user interfaces and interactive computer graphics. I will show examples in computer-aided design using physical simulation, generative models of images, 3D models and acoustic signals.īiography Takeo Igarashi is a professor in the Department of Creative Informatics at The University of Tokyo. In this talk, I will introduce methods for human intervention and control of such generative processes and other computer-assisted content generation systems. However, generative process using deep learning is a black box, which makes it difficult for humans to understand and control. Takeo Igarashi, Professor, The University of TokyoĪbstract Generative models that apply deep learning to the generation of contents such as images and sound are attracting attention. In addition, Luo served as an organizing or program committee member for numerous technical conferences sponsored by IEEE, ACM, AAAI, ACL, IAPR and SPIE, including, most notably, program co-chair of the 2010 ACM Multimedia Conference, 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016 ACM Conference on Multimedia Retrieval and 2017 IEEE International Conference on Image Processing.Ģ0 - Monday - 4:10 p.m. He is also an active member of the research community: a fellow of NAI, Association for Computing Machinery (ACM), Association for the Advancement of Artificial Intelligence (AAAI), Institute of Electrical and Electronics Engineers (IEEE), International Association for Pattern Recognition, SPIE, editor-in-chief of the IEEE Transactions on Multimedia (2020-2022), as well as a member of the editorial boards of the IEEE Transactions on Pattern Analysis and Machine Intelligence (2006-2011), IEEE Transactions on Multimedia (2004-2009, 2013-2016), IEEE Transactions on Circuits and Systems for Video Technology (2010-2012), IEEE Transactions on Big Data (2018-present), Pattern Recognition (2002-2020), ACM Transactions on Intelligent Systems and Technology (2015-present) and so on. Luo has authored nearly 600 papers and over 90 U.S. His research focuses on computer vision, natural language processing, machine learning, data mining, social media, computational social science, and digital health. We propose federated learning methods to alleviate the problem with applications in molecular graphs, medical images, and time series data.īiography Jiebo Luo is the Albert Arendt Hopeman Professor of Engineering and professor of computer science at the University of Rochester. Second, due to privacy concerns and transmission load, directly transferring data from edge devices to a centralized server to train a unified model is usually infeasible, while the data are often heterogeneously distributed among different insolated edge devices. We adopt adversarial learning to obtain bias/domain invariant features so that the learned model can generalize to out-of-distribution testing data. First, real-world data inevitably contains noise and bias. This talk will cover several related research topics. Such challenging datasets may make the learned model unreliable and pose threats to the learned model's generalization capacity for unseen testing data. However, in real-life applications, the distribution of testing data is likely different from that of training data because of many factors, including distribution shift, domain shift, isolated data server, noisy labels, and so on. Jiebo Luo, Professor, University of RochesterĪbstract Data-driven deep models will inherit the characteristics of the training data and thus can handle the in-distribution testing data well. Learning to Generalize to Out-of-Distribution Data Dr. 2022-2023 Distinguished Lecture Series March 2023Ħ - Monday - 4:10 p.m. This distinguished lecture series is sponsored by academia and industry. Each year the Department of Computer Science and Engineering invites prominent computer scientists and computer engineers to visit Texas A&M and speak about their research.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |