Negative to Positive Co-learning with Aggressive Modality Dropout
This addresses performance degradation in multimodal learning for AI applications, though it appears incremental as it builds on existing co-learning techniques.
The paper tackles the problem of negative co-learning in multimodal models by using aggressive modality dropout, which reverses it to positive co-learning and improves accuracy by up to 20% in some experiments.
This paper aims to document an effective way to improve multimodal co-learning by using aggressive modality dropout. We find that by using aggressive modality dropout we are able to reverse negative co-learning (NCL) to positive co-learning (PCL). Aggressive modality dropout can be used to "prep" a multimodal model for unimodal deployment, and dramatically increases model performance during negative co-learning, where during some experiments we saw a 20% gain in accuracy. We also benchmark our modality dropout technique against PCL to show that our modality drop out technique improves co-learning during PCL, although it does not have as much as an substantial effect as it does during NCL. Github: https://github.com/nmagal/modality_drop_for_colearning