Enhancing Eye-Tracking Performance through Multi-Task Learning Transformer

TL;DR

This paper presents a novel EEG signal reconstruction sub-module integrated into deep learning models, significantly improving eye-tracking accuracy while maintaining end-to-end training and reducing pre-training costs.

Contribution

Introduces an unsupervised EEG reconstruction sub-module compatible with various models, enhancing feature extraction and performance in eye-tracking tasks within a multi-task learning framework.

Findings

Achieved RMSE of 54.1mm, outperforming existing methods.

Enhanced feature representation capabilities of deep models.

Reduced computational costs compared to pre-training autoencoders.

Abstract

In this study, we introduce an innovative EEG signal reconstruction sub-module designed to enhance the performance of deep learning models on EEG eye-tracking tasks. This sub-module can integrate with all Encoder-Classifier-based deep learning models and achieve end-to-end training within a multi-task learning framework. Additionally, as the module operates under unsupervised learning, it is versatile and applicable to various tasks. We demonstrate its effectiveness by incorporating it into advanced deep-learning models, including Transformers and pre-trained Transformers. Our results indicate a significant enhancement in feature representation capabilities, evidenced by a Root Mean Squared Error (RMSE) of 54.1mm. This represents a notable improvement over existing methods, showcasing the sub-module's potential in refining EEG-based model performance. The success of this approach suggests that this reconstruction sub-module is capable of enhancing the feature extraction ability of the encoder. Due to the sub-module being mounted as a sub-task under the main task and maintained through a multi-task learning framework, our model preserves the end-to-end training process of the original model. In contrast to pre-training methods like autoencoder, our model saves computational costs associated with pre-training and exhibits greater flexibility in adapting to various model structures. Benefiting from the unsupervised nature of the sub-module, it can be applied across diverse tasks. We believe it represents a novel paradigm for improving the performance of deep learning models in EEG-related challenges.