MLSP-L9.6
THE DOUBLE-EDGED SWORD OF AI SAFETY: BALANCING ANOMALY DETECTION AND OOD GENERALIZATION VIA MODEL ANCHORING
Vivek Narayanaswamy, Rushil Anirudh, Jayaraman J. Thiagarajan, Lawrence Livermore National Laboratory, United States of America
Session:
MLSP-L9: Robustness and Trustworthy Machine Learning I Lecture
Track:
Machine Learning for Signal Processing
Location:
Room E3
Presentation Time:
Wed, 17 Apr, 14:50 - 15:10 (UTC +9)
Session Co-Chairs:
Jayaraman J. Thiagarajan, Lawrence Livermore National Laboratory and Jennifer Williams, University of Southampton
Session MLSP-L9
MLSP-L9.1: UNDERSTANDING DATA AUGMENTATION FROM A ROBUSTNESS PERSPECTIVE
Zhendong Liu, Jie Zhang, Qiangqiang He, Chongjun Wang, Nanjing University, China
MLSP-L9.2: ON ESTIMATING LINK PREDICTION UNCERTAINTY USING STOCHASTIC CENTERING
Puja Trivedi, Danai Koutra, University of Michigan, United States of America; Jayaraman J. Thiagarajan, Lawrence Livermore National Laboratory, United States of America
MLSP-L9.3: DEEPGRE: GLOBAL ROBUSTNESS EVALUATION OF DEEP NEURAL NETWORKS
Tianle Zhang, Jiaxu Liu, Yanghao Zhang, Ronghui Mu, Wenjie Ruan, University of Liverpool, United Kingdom of Great Britain and Northern Ireland
MLSP-L9.4: Adversarial Robustness of Convolutional Models Learned in the Frequency Domain
Subhajit Chaudhury, IBM Research, United States of America; Toshihiko Yamasaki, The University of Tokyo, Japan
MLSP-L9.5: NOISE-RESISTANT GRAPH NEURAL NETWORK FOR NODE CLASSIFICATION
Zichao Deng, Nanyang Technological University and Alibaba Group, Singapore; Han Yu, Nanyang Technological University, Singapore
MLSP-L9.6: THE DOUBLE-EDGED SWORD OF AI SAFETY: BALANCING ANOMALY DETECTION AND OOD GENERALIZATION VIA MODEL ANCHORING
Vivek Narayanaswamy, Rushil Anirudh, Jayaraman J. Thiagarajan, Lawrence Livermore National Laboratory, United States of America
Contacts