Key result
Bayesian deep learning predicts emotional valence from heartbeat time series with 90% accuracy.
Why the study?
Real-world collection of physiological data is largely limited to wearable heartbeat time series, and emotion prediction applications require uncertainty quantification to guide downstream decision-making.
Does a Bayesian deep learning framework using unimodal heartbeat time series accurately predict emotional valence?
Does a Bayesian deep learning framework using unimodal heartbeat time series accurately predict emotional valence?
A Bayesian deep learning framework using unimodal heartbeat time series can accurately predict emotional valence with quantifiable uncertainty, enabling potential applications in continuous non-invasive emotion monitoring.
May support non-invasive emotion monitoring; leaves open clinical validation before adoption.
Automatic prediction of emotion promises to revolutionise human-computer interaction. Recent trends involve fusion of multiple data modalities <tex-math notation="LaTeX">$-$</tex-math><alternatives>-</alternatives> audio, visual, and physiological <tex-math notation="LaTeX">$-$</tex-math><alternatives>-</alternatives> to classify emotional state. However, in practice, collection of physiological data ‘in the wild’ is currently limited to heartbeat time series of the kind generated by affordable wearable heart monitors. Furthermore, real-world applications of emotion prediction often require some measure of uncertainty over model output, in order to inform downstream decision-making. We present here an end-to-end deep learning model for classifying emotional valence from unimodal heartbeat time series. We further propose a Bayesian framework for modelling uncertainty over these valence predictions, and describe a probabilistic procedure for choosing to accept or reject model output according to the intended application. We benchmarked our framework against two established datasets and achieved peak classification accuracy of 90 percent. These results lay the foundation for applications of affective computing in real-world domains such as healthcare, where a high premium is placed on non-invasive collection of data, and predictive certainty.
No takes yet. Share an insight, caveat, or question.
Harper et al. (2020) studied Healthy participants (emotion prediction) (n=65). Bayesian deep learning framework (LSTM+CNN) vs. Non-Bayesian models and previous benchmarks was evaluated on Classification accuracy of emotional valence (high/low). An end-to-end Bayesian deep learning framework achieved a peak classification accuracy of 90% for predicting emotional valence from unimodal heartbeat time series.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: