Discrete Cosine Transform Based Decorrelated Attention for Vision Transformers
This work addresses efficiency and training challenges for Vision Transformers in computer vision, though it is incremental as it builds on existing Transformer architectures.
The paper tackles the initialization problem in Vision Transformers by proposing a DCT-based initialization method that enhances accuracy in classification tasks, and introduces a DCT-based compression technique that reduces computational overhead while maintaining accuracy.
Central to the Transformer architectures' effectiveness is the self-attention mechanism, a function that maps queries, keys, and values into a high-dimensional vector space. However, training the attention weights of queries, keys, and values is non-trivial from a state of random initialization. In this paper, we propose two methods. (i) We first address the initialization problem of Vision Transformers by introducing a simple, yet highly innovative, initialization approach utilizing discrete cosine transform (DCT) coefficients. Our proposed DCT-based \textit{attention} initialization marks a significant gain compared to traditional initialization strategies; offering a robust foundation for the attention mechanism. Our experiments reveal that the DCT-based initialization enhances the accuracy of Vision Transformers in classification tasks. (ii) We also recognize that since DCT effectively decorrelates image information in the frequency domain, this decorrelation is useful for compression because it allows the quantization step to discard many of the higher-frequency components. Based on this observation, we propose a novel DCT-based compression technique for the attention function of Vision Transformers. Since high-frequency DCT coefficients usually correspond to noise, we truncate the high-frequency DCT components of the input patches. Our DCT-based compression reduces the size of weight matrices for queries, keys, and values. While maintaining the same level of accuracy, our DCT compressed Swin Transformers obtain a considerable decrease in the computational overhead.