CHUCKLE -- When Humans Teach AI To Learn Emotions The Easy Way
This work addresses emotion recognition, a subjective task, by improving training efficiency and robustness, though it is incremental as it builds on existing curriculum learning methods.
The paper tackles the problem of curriculum learning for emotion recognition by proposing CHUCKLE, a perception-driven framework that uses human annotator agreement to define sample difficulty, resulting in a 6.56% relative mean accuracy increase for LSTMs and 1.61% for Transformers over non-curriculum baselines while reducing gradient updates.
Curriculum learning (CL) structures training from simple to complex samples, facilitating progressive learning. However, existing CL approaches for emotion recognition often rely on heuristic, data-driven, or model-based definitions of sample difficulty, neglecting the difficulty for human perception, a critical factor in subjective tasks like emotion recognition. We propose CHUCKLE (Crowdsourced Human Understanding Curriculum for Knowledge Led Emotion Recognition), a perception-driven CL framework that leverages annotator agreement and alignment in crowd-sourced datasets to define sample difficulty, under the assumption that clips challenging for humans are similarly hard for machine learning models. Empirical results suggest that CHUCKLE increases the relative mean accuracy by 6.56% for LSTMs and 1.61% for Transformers over non-curriculum baselines, while reducing the number of gradient updates, thereby enhancing both training efficiency and model robustness.