In-Person Meeting Location:

Storey Innovation Center 1400

Live Meeting Link for Virtual Audience

Abstract: 

Multi-label image classification aims to predict all possible labels in an image. It is usually formulated as a partial-label learning problem, since it could be expensive in practice to annotate all the labels in every training image. Existing works on partial-label learning focus on the case where each training image is labeled with only a subset of its positive/negative labels. To effectively address partial-label classification, this paper proposes an end-to-end Generic Game-theoretic Network for Partial-label learning (G2NetPL), which can be applied to most partial-label settings, including a very challenging, but annotation-efficient case where only a subset of the training images are labeled, each with only one positive label, while the rest of the training images remain unlabeled. In G2NetPL, each unobserved label is associated with a soft pseudo label, which, together with the network, formulates a two-player non-zero-sum non-cooperative game. The objective of the network is to minimize the loss function with given pseudo labels, while the pseudo labels will seek convergence to 1 (positive) or 0 (negative) with a penalty of deviating from the predicted labels determined by the network. In addition, we introduce a confidence-aware scheduler into the loss of the network to adaptively perform easy-to-hard learning for different labels. Extensive experiments demonstrate that G2NetPL outperforms many state-of-the-art multi-label classification methods under various partial-label settings on three different datasets.

Speaker's Bio:

Xiaofeng Wang is associate professor in the Department of Electrical Engineering at the University of South Carolina (UofSC), Columbia. He earned his B.S. degree in Applied Mathematics and M.S. in Operation Research and Control Theory from East China Normal University, China, in 2000 and 2003, respectively, and obtained his PhD degree in Electrical Engineering at the University of Notre Dame in 2009. After that, he worked as postdoctoral research associate at the University of Illinois at Urbana-Champaign before he joined UofSC. His research interests include robotics and control, cyber-physical systems, autonomous systems, multi-agent systems, and machine learning. He is associate editor of IEEE Transactions on Aerospace and Electronic Systems, Journal of The Franklin Institute, and IEEE CSS conference board. He was the recipient of the best paper award in the Annual Conference of the Prognostics and Health Management Society in 2014 and the finalist of the best paper award in International Conference on Cyber-Physical Systems in 2013.

The Office of the Vice President for Research is honored to welcome UofSC faculty researchers Forest Agostinelli, Orgul Ozturk, Jane Roberts and Bryant Walker Smith, as panelists for AI: Game Changer or Game Over?

Event Overview

The rapid proliferation of artificial intelligence in the 21st century is both promising and fraught, and for good reason—for decades, popular culture has envisioned how this futuristic technology might serve or even destroy humanity. From Rosey the sassy robot maid in the Jetsons to the sinister HAL 9000 computer in Stanley Kubrick’s 2001: A Space Odyssey, and including seemingly endless depictions in-between those extremes, our art has anticipated both helpful, symbiotic relationships and destructive confrontations between biological humanity and human-created intelligent technologies. But now that AI is here with us, what is the reality? How are artificial intelligences serving humanity today and how will their roles evolve tomorrow? What pitfalls come with the benefits of using AI? How do we harness the power of AI without becoming dangerously over-reliant?

These are some of the questions the Office of the Vice President for Research will invite the university community to explore on Thursday, October 6, 2022, when we convene a panel of university faculty experts to discuss their insights on the ethics and implications of artificial intelligence.