MLSP-L26: Explainable Machine Learning for Signal Processing
Oral
Thu, 7 May, 16:30 - 18:30
Location: Room 113
Session Type: Oral
Track: Machine Learning for Signal Processing [ML]
Click the to view the manuscript on IEEE Xplore Open Preview
Thu, 7 May, 16:30 - 16:50

MLSP-L26.1: PREIG: Physics-informed and Reinforcement-driven Interpretable GRU for Commodity Demand Forecasting
Hongwei Ma, Junbin Gao, Minh-Ngoc Tran, Jiayu Fang, The University of Sydney, Australia
Thu, 7 May, 16:50 - 17:10

MLSP-L26.2: GENFACTS-GENERATIVE COUNTERFACTUAL EXPLANATIONS FOR MULTI-VARIATE TIME SERIES
Sarah Seifi, Anass Ibrahimi, Technical University Munich/Infineon Technologies AG, Germany; Tobias Sukianto, Cecilia Carbonelli, Infineon Technologies AG, Germany; Lorenzo Servadei, Robert Wille, Technical University Munich, Germany
Thu, 7 May, 17:10 - 17:30

MLSP-L26.3: GRAM-SCHMIDT FEATURE SELECTION FOR CLASS ACTIVATION MAPS
Kaveh Safavigerdini, University of Missouri-Columbia, United States of America; Bahram Yaghooti, Bruno Sinopoli, Washington University in St. Louis, United States of America; Kannappan Palaniappan, University of Missouri-Columbia, United States of America
Thu, 7 May, 17:30 - 17:50

MLSP-L26.4: FIG: FREQUENCY-BASED INTEGRATED GRADIENTS FOR ROBUST FEATURE ATTRIBUTION
Kun Yang, Southwest University, China; Zhizhong Niu, Guizhou University of Commerce, China; Fangmin Xie, Ziyao Tang, Guoqiang Xiao, Southwest University, China
Thu, 7 May, 17:50 - 18:10

MLSP-L26.5: GEODESIC PROTOTYPE MATCHING VIA DIFFUSION MAPS FOR INTERPRETABLE FINE-GRAINED RECOGNITION
Junhao Jia, Yunyou Liu, Hangzhou Dianzi University, China; Yifei Sun, Huangwei Chen, Zhejiang University, China; Feiwei Qin, Hangzhou Dianzi University, China; Changmiao Wang, Shenzhen Research Institute of Big Data, China; Yong Peng, Hangzhou Dianzi University, China
Thu, 7 May, 18:10 - 18:30

MLSP-L26.6: AR&D: A Framework for Retrieving and Describing Concepts for Interpreting AudioLLMs
Townim Faisal Chowdhury, Ta Duc Huy, Australian Institute for Machine Learning, University of Adelaide, Australia, Australia; Siqi Pan, Jeremy Stoddard, Dolby Laboratories, Australia, Australia; Zhibin Liao, School of Computer and Mathematical Sciences, University of Adelaide, Australia, Australia