AffecTech Researchers Honoured at International Conference on Machine Learning, Optimization, and Data Science

 

AffecTech researchers Andrea Patanè and Professor Marta Kwiatkowska have received the prestigious best paper award at LOD 2018: the International Conference on Machine Learning, Optimization, and Data Science, for their work which advances our understanding of affective analysis of physiological signals.

The 4th Annual Conference on Machine Learning, Optimization and Data science (LOD) is a single-track machine learning, computational optimization, data science conference.

Calibrating the Classifier: Siamese Neural Network Architecture for End-to-End Arousal Recognition from ECG” won the LOD 2018 Best Paper Award at the conference that took place September 13-16, 2018 in Volterra, Tuscany, Italy.

The recognition for the paper is based on its innovative research into emotion recognition in mobile wearable devices, described by the conference leadership as “visionary”.

Authors AffecTech researcher Andrea Patanè and AffecTech academic Professor Marta Kwiatkowska, are both based at the Department of Computer Science, University of Oxford, UK, with Andrea working exclusively on the EU funded AffecTech research project – created to develop new personal technologies for affective health, ie the relatively common mental health conditions of depression, anxiety and bipolar.

Their paper presents a deep learning framework for arousal recognition from ECG (electrocardiogram) signals, and specifically designs an end-to-end convolutional and recurrent neural network architecture to extract features from ECG, that advances existing technologies.

The abstract notes that: “The key novelty is our use of a shared parameter siamese architecture to implement user-specific feature calibration. At each forward and backward pass, we concatenate to the input a user-dependent template that is processed by an identical copy of the network.

“The siamese architecture makes feature calibration an integral part of the training process, allowing modelling of general dependencies between the user’s ECG at rest and those during emotion elicitation. On leave-one-user-out cross validation, the proposed architecture obtains +21.5% score increase compared to state-of-the-art techniques. Comparison with alternative network architectures demonstrates the effectiveness of the siamese network in achieving user-specific feature calibration.”

The global scientific, medical publisher Springer sponsored the LOD 2018 Best Paper Award with a prize of EUR 1,000.

The LOD Conference Manifesto cites that: “The problem of understanding intelligence is said to be the greatest problem in science today… Arguably, the problem of learning represents a gateway to understanding intelligence in brains and machines, to discovering how the human brain works, and to making intelligent machines…”

About AffecTech

AffecTech is established with support from the Marie Skłodowska-Curie Innovative Training Network (ITN) via the European Commission’s Horizon 2020 (H2020) research and innovation programme. AffecTech is a European Innovative Training Network funded under Marie Skłodowska-Curie grant agreement No 722022.