# Take it Personally: The Limits of General SSL Representations for Real-Life PPG Emotion Detection

arXiv: https://arxiv.org/abs/2608.14675 · HTML: https://mlsanity.com/paper/2608.14675/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 37 references that are in the map.
- First posted on arXiv 3 Aug 2026.

## Builds on (most cited first, 7 of 7)

- [Time-Series Representation Learning via Temporal and Contextual Contrasting](https://mlsanity.com/paper/2106.14112/) (2021)
- [CLOCS: Contrastive Learning of Cardiac Signals Across Space, Time, and Patients](https://mlsanity.com/paper/2005.13249/) (2020)
- [Self-supervised ECG Representation Learning for Emotion Recognition](https://mlsanity.com/paper/2002.03898/) (2020)
- [Utilizing Deep Learning Towards Multi-modal Bio-sensing and Vision-based Affective Computing](https://mlsanity.com/paper/1905.07039/) (2019)
- [Transformer-Based Self-Supervised Learning for Emotion Recognition](https://mlsanity.com/paper/2204.05103/) (2022)
- [A Bayesian Deep Learning Framework for End-To-End Prediction of Emotion from Heartbeat](https://mlsanity.com/paper/1902.03043/) (2019)
- [Scaling Representation Learning from Ubiquitous ECG with State-Space Models](https://mlsanity.com/paper/2309.15292/) (2023)

## Abstract (arXiv)

While Self-Supervised Learning (SSL) effectively extracts general representations from noisy, unconstrained physiological signals such as photoplethysmography (PPG), its suitability for highly subjective tasks remains unproven. In this work, we evaluate the efficacy of PPG-based SSL for real-life intense emotion detection. First, we pretrain a Real-Life PPG encoder (RL-PPG) on unconstrained, real-life data. As a rigorous sanity check, we demonstrate that these representations transfer exceptionally well to an objective physical activity recognition task, yielding almost 5-fold increase in performance over baselines in a leave-one-subject-out evaluation (LOSO). However, when applied to a~subjective real-life emotion detection task, these same general representations fail to surpass naive baselines under the LOSO protocol. Using an Across-Time validation strategy, we establish that incorporating an individual's personal data during fine-tuning is the main driver of predictive performance, outweighing the benefits of population-level pretraining. Ultimately, our findings indicate that in the evaluated scenario, general SSL representations may be insufficient for subjective affective inference, suggesting that personalization is likely a key component for real-world emotion recognition. To support future research, we share the code and pretrained RL-PPG~encoder~weights.