TY - GEN
T1 - Transfer Bayesian Meta-Learning Via Weighted Free Energy Minimization
AU - Zhang, Yunchuan
AU - Jose, Sharu Theresa
AU - Simeone, Osvaldo
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021
Y1 - 2021
N2 - Meta-learning optimizes the hyperparameters of a training procedure, such as its initialization, kernel, or learning rate, based on data sampled from a number of auxiliary tasks. A key underlying assumption is that the auxiliary tasks-known as meta-Training tasks-share the same generating distribution as the tasks to be encountered at deployment time-known as meta-Test tasks. This may, however, not be the case when the test environment differ from the meta-Training conditions. To address shifts in task generating distribution between meta-Training and meta-Testing phases, this paper introduces weighted free energy minimization (WFEM) for transfer meta-learning. We instantiate the proposed approach for non-parametric Bayesian regression and classification via Gaussian Processes (GPs). The method is validated on a toy sinusoidal regression problem, as well as on classification using miniImagenet and CUB data sets, through comparison with standard meta-learning of GP priors as implemented by PACOH.
AB - Meta-learning optimizes the hyperparameters of a training procedure, such as its initialization, kernel, or learning rate, based on data sampled from a number of auxiliary tasks. A key underlying assumption is that the auxiliary tasks-known as meta-Training tasks-share the same generating distribution as the tasks to be encountered at deployment time-known as meta-Test tasks. This may, however, not be the case when the test environment differ from the meta-Training conditions. To address shifts in task generating distribution between meta-Training and meta-Testing phases, this paper introduces weighted free energy minimization (WFEM) for transfer meta-learning. We instantiate the proposed approach for non-parametric Bayesian regression and classification via Gaussian Processes (GPs). The method is validated on a toy sinusoidal regression problem, as well as on classification using miniImagenet and CUB data sets, through comparison with standard meta-learning of GP priors as implemented by PACOH.
KW - Bayesian learning
KW - Gaussian Process
KW - Transfer Meta-learning
UR - http://www.scopus.com/inward/record.url?scp=85122807350&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85122807350&partnerID=8YFLogxK
U2 - 10.1109/MLSP52302.2021.9596239
DO - 10.1109/MLSP52302.2021.9596239
M3 - Conference contribution
AN - SCOPUS:85122807350
T3 - IEEE International Workshop on Machine Learning for Signal Processing, MLSP
BT - 2021 IEEE 31st International Workshop on Machine Learning for Signal Processing, MLSP 2021
PB - IEEE Computer Society
T2 - 31st IEEE International Workshop on Machine Learning for Signal Processing, MLSP 2021
Y2 - 25 October 2021 through 28 October 2021
ER -