Calibrated Multimodal Representation Learning with Missing Modalities
By: Xiaohao Liu , Xiaobo Xia , Jiaheng Wei and more
Potential Business Impact:
Helps computers learn from mixed-up, incomplete data.
Multimodal representation learning harmonizes distinct modalities by aligning them into a unified latent space. Recent research generalizes traditional cross-modal alignment to produce enhanced multimodal synergy but requires all modalities to be present for a common instance, making it challenging to utilize prevalent datasets with missing modalities. We provide theoretical insights into this issue from an anchor shift perspective. Observed modalities are aligned with a local anchor that deviates from the optimal one when all modalities are present, resulting in an inevitable shift. To address this, we propose CalMRL for multimodal representation learning to calibrate incomplete alignments caused by missing modalities. Specifically, CalMRL leverages the priors and the inherent connections among modalities to model the imputation for the missing ones at the representation level. To resolve the optimization dilemma, we employ a bi-step learning method with the closed-form solution of the posterior distribution of shared latents. We validate its mitigation of anchor shift and convergence with theoretical guidance. By equipping the calibrated alignment with the existing advanced method, we offer new flexibility to absorb data with missing modalities, which is originally unattainable. Extensive experiments and comprehensive analyses demonstrate the superiority of CalMRL. Our code, model checkpoints, and evaluation raw data will be publicly available.
Similar Papers
MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates
CV and Pattern Recognition
Fixes AI when some information is missing.
MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates
CV and Pattern Recognition
Helps computers learn from mixed-up information.
Causal Debiasing Medical Multimodal Representation Learning with Missing Modalities
Machine Learning (CS)
Fixes medical AI when data is missing.