Reconsidering Representation Alignment for Multi-view Clustering
This work addresses multi-view clustering for researchers, showing that avoiding alignment can be beneficial, but it is incremental as it builds on existing methods with a new baseline and enhancement.
The paper tackles the problem of representation alignment in multi-view clustering by identifying drawbacks that reduce cluster separability and view prioritization, and develops a baseline model without alignment that matches or exceeds state-of-the-art performance, with a contrastive learning addition improving results by a large margin on several datasets.
Aligning distributions of view representations is a core component of today's state of the art models for deep multi-view clustering. However, we identify several drawbacks with naïvely aligning representation distributions. We demonstrate that these drawbacks both lead to less separable clusters in the representation space, and inhibit the model's ability to prioritize views. Based on these observations, we develop a simple baseline model for deep multi-view clustering. Our baseline model avoids representation alignment altogether, while performing similar to, or better than, the current state of the art. We also expand our baseline model by adding a contrastive learning component. This introduces a selective alignment procedure that preserves the model's ability to prioritize views. Our experiments show that the contrastive learning component enhances the baseline model, improving on the current state of the art by a large margin on several datasets.