In this paper, we consider the challenging problem of multi-source zero shot domain generalization (MDG), where labeled training data from multiple source domains are available but with no access to data from the target domain. Many methods have been proposed to address this problem, but surprisingly the naiive solution of pooling all source data together and training a single ERM model is highly competitive. Constructing an ensemble of deep classifiers is a popular approach for building models that are calibrated under challenging distribution shifts. Hence, we propose MulDEns (Multi-Domain Deep Ensembles), a new approach for constructing deep ensembles in multi-domain problems that does not require to construct domain-specific models. Our empirical studies on multiple standard benchmarks show that MulDEns significantly outperforms ERM and existing ensembling solutions for MDG.
Kowshik Thopalli (Arizona State University)
Sameeksha Katoch (Arizona State University)
Doctoral student in Electrical Engineering at Arizona State University working with focus on Machine Learning, Computer Vision and Signal Processing.
Jayaraman Thiagarajan (Lawrence Livermore National Laboratory)
Pavan Turaga (Arizona State University)
Andreas Spanias (ASU)
Andreas Spanias is Professor in the School of Electrical, Computer, and Energy Engineering at Arizona State University (ASU). He is also the director of the Sensor Signal and Information Processing (SenSIP) center and the founder of the SenSIP industry consortium (also an NSF I/UCRC site). His research interests are in the areas of adaptive signal processing, speech processing, machine learning and sensor systems. He and his student team developed the computer simulation software Java-DSP and its award-winning iPhone/iPad and Android versions. He is author of two textbooks: Audio Processing and Coding by Wiley and DSP; An Interactive Approach (2nd Ed.). He contributed to more than 350 papers, 11 monographs 11 full patents, 10 provisional patents and 12 patent pre-disclosures. He served as Associate Editor of the IEEE Transactions on Signal Processing and as General Co-chair of IEEE ICASSP-99. He also served as the IEEE Signal Processing Vice-President for Conferences. Andreas Spanias is co-recipient of the 2002 IEEE Donald G. Fink paper prize award and was elected Fellow of the IEEE in 2003. He served as Distinguished Lecturer for the IEEE Signal processing society in 2004. He is a series editor for the Morgan and Claypool lecture series on algorithms and software. He received the 2018 IEEE Phoenix Chapter award with citation: “For significant innovations and patents in signal processing for sensor systems.” He also received the 2018 IEEE Region 6 Outstanding Educator Award (across 12 states) with citation: “For outstanding research and education contributions in signal processing.” He was elected recently to Senior Member of the National Academy of Inventors (NAI).
More from the Same Authors
2021 : Data Efficient Domain Adaptation using FiLM »
Sinjini Mitra · Rushil Anirudh · Jayaraman Thiagarajan · Pavan Turaga
2021 : Re-labeling Domains Improves Multi-Domain Generalization »
Kowshik Thopalli · Pavan Turaga · Jayaraman Thiagarajan
2021 : Unsupervised Attribute Alignment for Characterizing Distribution Shift »
Matthew Olson · Rushil Anirudh · Jayaraman Thiagarajan · Timo Bremer · Weng-Keen Wong · Shusen Liu
2021 : Geometric Priors for Scientific Generative Models in Inertial Confinement Fusion »
Ankita Shukla · Rushil Anirudh · Eugene Kur · Jayaraman Thiagarajan · Timo Bremer · Brian K Spears · Tammy Ma · Pavan Turaga
2022 : Learning Free Energy Pathways through Reinforcement Learning of Adaptive Steered Molecular Dynamics »
Nicholas Ho · John Kevin Cava · John Vant · Ankita Shukla · Jacob Miratsky · Pavan Turaga · Ross Maciejewski · Abhishek Singharoy
2021 Poster: Designing Counterfactual Generators using Deep Model Inversion »
Jayaraman Thiagarajan · Vivek Sivaraman Narayanaswamy · Deepta Rajan · Jia Liang · Akshay Chaudhari · Andreas Spanias