EpilepsyGAN: Synthetic Epileptic Brain Activities With Privacy Preservation.

Epilepsy is a chronic neurological disorder affecting more than 65 million people worldwide and manifested by recurrent unprovoked seizures. The unpredictability of seizures not only degrades the quality of life of the patients, but it can also be life-threatening. Modern systems monitoring electroencephalography (EEG) signals are being currently developed with the view to detect epileptic seizures in order to alert caregivers and reduce the impact of seizures on patients' quality of life. Such seizure detection systems employ state-of-the-art machine learning algorithms that require a large amount of labeled personal data for training. However, acquiring EEG signals during epileptic seizures is a costly and time-consuming process for medical experts and patients. Furthermore, this data often contains sensitive personal information, presenting privacy concerns. In this work, we generate synthetic seizure-like brain electrical activities, i.e., EEG signals, that can be used to train seizure detection algorithms, alleviating the need for sensitive recorded data. Our experiments show that the synthetic seizure data generated with our GAN model succeeds at preserving the privacy of the patients without producing any degradation in performance during seizure monitoring.

Recognition and Repetition Counting for ComplexPhysical Exercises with Deep Learning.

Activity recognition using off-the-shelf smartwatches is an important problem in humanactivity recognition. In this paper, we present an end-to-end deep learning approach, able to provideprobability distributions over activities from raw sensor data. We apply our methods to 10 complexfull-body exercises typical in CrossFit, and achieve a classification accuracy of 99.96%. We additionallyshow that the same neural network used for exercise recognition can also be used in repetitioncounting. To the best of our knowledge, our approach to repetition counting is novel and performswell, counting correctly within an error of 1 repetitions in 91% of the performed sets.

Scalable funding of Bitcoin micropayment channel networks.

The Bitcoin network has scalability problems. To increase its transaction rate and speed, micropayment channel networks have been proposed; however, these require to lock funds into specific channels. Moreover, the available space in the blockchain does not allow scaling to a worldwide payment system. We propose a new layer that sits in between the blockchain and the payment channels. The new layer addresses the scalability problem by enabling trustless off-blockchain channel funding. It consists of shared accounts of groups of nodes that flexibly create one-to-one channels for the payment network. The new system allows rapid changes of the allocation of funds to channels and reduces the cost of opening new channels. Instead of one blockchain transaction per channel, each user only needs one transaction to enter a group of nodes-within the group the user can create arbitrarily many channels. For a group of 20 users with 100 intra-group channels, the cost of the blockchain transactions is reduced by 90% compared to 100 regular micropayment channels opened on the blockchain. This can be increased further to 96% if Bitcoin introduces Schnorr signatures with signature aggregation.

Distributed algorithms for sensor networks.

Distributed algorithms are an established tool for designing protocols for sensor networks. In this paper, we discuss the relation between distributed computing theory and sensor network applications. We also present a few basic and illustrative distributed algorithms.