Affective
Intelligence Lab
Understanding humans.
We explore multimodal machine learning and affective computing.
We study how machine learning can help us understand emotion and behavior from facial expressions, voice, and language. By considering these signals together and in context, we explore what models can reveal about human experience, and where their interpretations fall short.
Recent publications
-
Shift-ACT: Test-time adaptation via shift-aware class-specific dynamic thresholding for domain generalization
Read paper -
AdaPAC: Prototypical anchored contrastive test time adaptation for domain generalization
Read paper -
Affect and Personality Aided Modeling of Transcribed Speech for Depression Severity Estimation
Read paper