FingerFlex: Inferring Finger Trajectories from ECoG signals

Motor brain-computer interface (BCI) development relies critically on neural time series decoding algorithms. Recent advances in deep learning architectures allow for automatic feature selection to approximate higher-order dependencies in data. This article presents the FingerFlex model - a convolutional encoder-decoder architecture adapted for finger movement regression on electrocorticographic (ECoG) brain data. State-of-the-art performance was achieved on a publicly available BCI competition IV dataset 4 with a correlation coefficient between true and predicted trajectories up to 0.74. The presented method provides the opportunity for developing fully-functional high-precision cortical motor brain-computer interfaces.

Few-shot Generation of Personalized Neural Surrogates for Cardiac Simulation via Bayesian Meta-Learning

Clinical adoption of personalized virtual heart simulations faces challenges in model personalization and expensive computation. While an ideal solution is an efficient neural surrogate that at the same time is personalized to an individual subject, the state-of-the-art is either concerned with personalizing an expensive simulation model, or learning an efficient yet generic surrogate. This paper presents a completely new concept to achieve personalized neural surrogates in a single coherent framework of meta-learning (metaPNS). Instead of learning a single neural surrogate, we pursue the process of learning a personalized neural surrogate using a small amount of context data from a subject, in a novel formulation of few-shot generative modeling underpinned by: 1) a set-conditioned neural surrogate for cardiac simulation that, conditioned on subject-specific context data, learns to generate query simulations not included in the context set, and 2) a meta-model of amortized variational inference that learns to condition the neural surrogate via simple feed-forward embedding of context data. As test time, metaPNS delivers a personalized neural surrogate by fast feed-forward embedding of a small and flexible number of data available from an individual, achieving -- for the first time -- personalization and surrogate construction for expensive simulations in one end-to-end learning framework. Synthetic and real-data experiments demonstrated that metaPNS was able to improve personalization and predictive accuracy in comparison to conventionally-optimized cardiac simulation models, at a fraction of computation.

Thunderstorm nowcasting with deep learning: a multi-hazard data fusion model

Predictions of thunderstorm-related hazards are needed in several sectors, including first responders, infrastructure management and aviation. To address this need, we present a deep learning model that can be adapted to different hazard types. The model can utilize multiple data sources; we use data from weather radar, lightning detection, satellite visible/infrared imagery, numerical weather prediction and digital elevation models. It can be trained to operate with any combination of these sources, such that predictions can still be provided if one or more of the sources become unavailable. We demonstrate the ability of the model to predict lightning, hail and heavy precipitation probabilistically on a 1 km resolution grid, with a time resolution of 5 min and lead times up to 60 min. Shapley values quantify the importance of the different data sources, showing that the weather radar products are the most important predictors for all three hazard types.

Adversarial Guitar Amplifier Modelling With Unpaired Data

We propose an audio effects processing framework that learns to emulate a target electric guitar tone from a recording. We train a deep neural network using an adversarial approach, with the goal of transforming the timbre of a guitar, into the timbre of another guitar after audio effects processing has been applied, for example, by a guitar amplifier. The model training requires no paired data, and the resulting model emulates the target timbre well whilst being capable of real-time processing on a modern personal computer. To verify our approach we present two experiments, one which carries out unpaired training using paired data, allowing us to monitor training via objective metrics, and another that uses fully unpaired data, corresponding to a realistic scenario where a user wants to emulate a guitar timbre only using audio data from a recording. Our listening test results confirm that the models are perceptually convincing.

A Location-Based Global Authorization Method for Underwater Security

National or international maritime authorities are used to handle requests for licenses for all kinds of marine activities. These licenses constitute authorizations limited in time and space, but there is no technical security service to check for the authorization of a wide range of marine assets. We have noted secure AIS solutions suitable for more or less constantly internet-connected assets such as ships with satellite connections. The additional constraints posed by underwater autonomous assets, namely less power and connectivity, can be mitigated by using symmetric cryptography. We propose a security service that allows the automatized check of asset authorization status based on large symmetric keys. Key generation can take place at a central authority according to the time and space limitations of a license, i.e. timestamped and geocoded. Our solution harnesses the exceptionally large key size of the RC5 cipher and the standardized encoding of geocells in the Open Location Code system. While we developed and described our solution for offshore underwater use, aerial and terrestrial environments could also make use of it if they are similarly bandwidth constrained or want to rely on quantum-resistant and computationally economic symmetric methods.

Personalizing Sustainable Agriculture with Causal Machine Learning

To fight climate change and accommodate the increasing population, global crop production has to be strengthened. To achieve the "sustainable intensification" of agriculture, transforming it from carbon emitter to carbon sink is a priority, and understanding the environmental impact of agricultural management practices is a fundamental prerequisite to that. At the same time, the global agricultural landscape is deeply heterogeneous, with differences in climate, soil, and land use inducing variations in how agricultural systems respond to farmer actions. The "personalization" of sustainable agriculture with the provision of locally adapted management advice is thus a necessary condition for the efficient uplift of green metrics, and an integral development in imminent policies. Here, we formulate personalized sustainable agriculture as a Conditional Average Treatment Effect estimation task and use Causal Machine Learning for tackling it. Leveraging climate data, land use information and employing Double Machine Learning, we estimate the heterogeneous effect of sustainable practices on the field-level Soil Organic Carbon content in Lithuania. We thus provide a data-driven perspective for targeting sustainable practices and effectively expanding the global carbon sink.

Leveraging Haptic Feedback to Improve Data Quality and Quantity for Deep Imitation Learning Models

Learning from demonstration (LfD) is a proven technique to teach robots new skills. Data quality and quantity play a critical role in LfD trained model performance. In this paper we analyze the effect of enhancing an existing teleoperation data collection system with real-time haptic feedback; we observe improvements in the collected data throughput and its quality for model training. Our experiment testbed was a mobile manipulator robot that opened doors with latch handles. Evaluation of teleoperated data collection on eight real world conference room doors found that adding the haptic feedback improved the data throughput by 6%. We additionally used the collected data to train six image-based deep imitation learning models, three with haptic feedback and three without it. These models were used to implement autonomous door-opening with the same type of robot used during data collection. Our results show that a policy from a behavior cloning model trained with haptic data performed on average 11% better than its counterpart with no haptic feedback data, indicating that haptic feedback resulted in collection of a higher quality dataset.

Deep Learning for Global Wildfire Forecasting

Climate change is expected to aggravate wildfire activity through the exacerbation of fire weather. Improving our capabilities to anticipate wildfires on a global scale is of uttermost importance for mitigating their negative effects. In this work, we create a global fire dataset and demonstrate a prototype for predicting the presence of global burned areas on a sub-seasonal scale with the use of segmentation deep learning models. Particularly, we present an open-access global analysis-ready datacube, which contains a variety of variables related to the seasonal and sub-seasonal fire drivers (climate, vegetation, oceanic indices, human-related variables), as well as the historical burned areas and wildfire emissions for 2001-2021. We train a deep learning model, which treats global wildfire forecasting as an image segmentation task and skillfully predicts the presence of burned areas 8, 16, 32 and 64 days ahead of time. Our work motivates the use of deep learning for global burned area forecasting and paves the way towards improved anticipation of global wildfire patterns.

Prompt-based Text Entailment for Low-Resource Named Entity Recognition

Pre-trained Language Models (PLMs) have been applied in NLP tasks and achieve promising results. Nevertheless, the fine-tuning procedure needs labeled data of the target domain, making it difficult to learn in low-resource and non-trivial labeled scenarios. To address these challenges, we propose Prompt-based Text Entailment (PTE) for low-resource named entity recognition, which better leverages knowledge in the PLMs. We first reformulate named entity recognition as the text entailment task. The original sentence with entity type-specific prompts is fed into PLMs to get entailment scores for each candidate. The entity type with the top score is then selected as final label. Then, we inject tagging labels into prompts and treat words as basic units instead of n-gram spans to reduce time complexity in generating candidates by n-grams enumeration. Experimental results demonstrate that the proposed method PTE achieves competitive performance on the CoNLL03 dataset, and better than fine-tuned counterparts on the MIT Movie and Few-NERD dataset in low-resource settings.

Multivariate Wasserstein Functional Connectivity for Autism Screening

Most approaches to the estimation of brain functional connectivity from the functional magnetic resonance imaging (fMRI) data rely on computing some measure of statistical dependence, or more generally, a distance between univariate representative time series of regions of interest (ROIs) consisting of multiple voxels. However, summarizing a ROI's multiple time series with its mean or the first principal component (1PC) may result to the loss of information as, for example, 1PC explains only a small fraction of variance of the multivariate signal of the neuronal activity. We propose to compare ROIs directly, without the use of representative time series, defining a new measure of multivariate connectivity between ROIs, not necessarily consisting of the same number of voxels, based on the Wasserstein distance. We assess the proposed Wasserstein functional connectivity measure on the autism screening task, demonstrating its superiority over commonly used univariate and multivariate functional connectivity measures.